medicare spending per beneficiary – post-acute care ... · acumen, llc. 3 . 1 overview . project...

Medicare Spending Per Beneficiary –

Post-Acute Care Measures

Public Comment Summary Report

March 2016

Acumen, LLC 500 Airport Blvd., Suite 365 Burlingame, CA 94010

2 Acumen, LLC | MSPB-PAC Public Comment Summary Report

TABLE OF CONTENTS 1 Overview .................................................................................................................................. 3 2 Stakeholder Comments: General and Measure-Specific .................................................... 6

2.1 Overall Approach and Process ......................................................................................... 6 2.1.1 Measure Intent ..................................................................................................... 6 2.1.2 Measure Development Process .......................................................................... 10 2.1.3 Implementation .................................................................................................. 13

2.2 Episode Construction ..................................................................................................... 14 2.2.1 Opening (Triggering) Episodes ......................................................................... 14 2.2.2 Defining the Episode Window ........................................................................... 16 2.2.3 Defining Treatment Services ............................................................................. 18 2.2.4 Defining Associated Services ............................................................................ 18 2.2.5 Excluding Clinically Unrelated Services ........................................................... 21 2.2.6 Closing Episodes................................................................................................ 23

2.3 Measure Calculation ...................................................................................................... 23 2.3.1 Implementing Episode-Level Exclusions .......................................................... 23 2.3.2 Risk Adjustment Approach ................................................................................ 24 2.3.3 MSPB-PAC Measure Calculation ..................................................................... 30

3 Overall Analysis and Recommendations ............................................................................ 34 3.1 Preliminary Recommendations ...................................................................................... 34 3.2 Overall Analysis of the Comments and Recommendations .......................................... 34

Appendix A . Public Comment Verbatim Report .................................................................... 35 A.1 Index of Public Comments Received............................................................................. 35 A.2 Verbatim Comments and Responses ............................................................................. 38

LIST OF TABLES Table A1. Index of Public Comments Received ........................................................................... 35

MSPB-PAC Public Comment Summary Report | Acumen, LLC 3

1 OVERVIEW

Project Title Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) Resource Use

Measures

Dates

• The Call for Public Comment ran from January 13 to 27, 2016. This was extended twice to January 29 and February 5, 2016.

• The Public Comment Summary Report was submitted to the Centers for Medicare & Medicaid Services (CMS) on February 19, 2016.

Project Overview

CMS has contracted with Acumen, LLC to develop the MSPB-PAC measures under the Calculating Episode-Based Costs from the Medicare Episode Grouper for Physician Feedback contract (HHSM-500-2011-000121, Task Order HHSM-500-T0008). As part of its measure development process, CMS has requested interested parties to submit comments on the candidate or concept measures that may be suitable for this project.

Project Objectives

The goal of this project is to develop resource use measures for PAC settings as mandated by the Improving Post-Acute Care Transformation Act of 2014 (IMPACT Act). These measures apply to skilled nursing facilities (SNFs), home health agencies (HHAs), long-term care hospitals (LTCHs), and inpatient rehabilitation facilities (IRFs) and will be reported in each respective PAC setting’s quality reporting program (QRP). The proposed MSPB-PAC episode-based measures will provide actionable and transparent information to support PAC providers’ efforts to promote care coordination and deliver high quality care at a lower cost to Medicare. A given PAC provider’s risk-adjusted Medicare spending is evaluated relative to that of the national median PAC provider in the same setting.

Proposed Measures

We proposed the following episode-based resource use measures for the PAC settings:

(1) Medicare Spending Per Beneficiary – Post-Acute Care (PAC) Skilled Nursing Facility Measure

(2) Medicare Spending Per Beneficiary – Post-Acute Care (PAC) Home Health Measure

(3) Medicare Spending Per Beneficiary – Post-Acute Care (PAC) Long-Term Care Hospital Measure

(4) Medicare Spending Per Beneficiary – Post-Acute Care (PAC) Inpatient Rehabilitation Facility Measure


Information about the Comments Received

• Public comments were solicited by:

o Public posting on CMS Public Comment website

o Email notification of Division of Community Post-Acute Care (DCPAC) stakeholders

o Email notification of MSPB-PAC Technical Expert Panel (TEP) members

• We received 45 responses regarding the MSPB-PAC measures via email during the public comment period.

o An index of the comments is at Appendix A, Table A1. Index of Public Comments Received.

o Verbatim comments are at Appendix A, Section A.2. The same information is also presented in table format as a separate attachment to this document, titled “Medicare Spending Per Beneficiary – Post-Acute Care Measures: Public Comment Verbatim Report”.

o We received comments from the following organizations:

Alliance for Home Health Quality and Innovation

Amedisys, Inc

American Academy of Physical Medicine and Rehabilitation

American Health Care Association

American Hospital Association

American Physical Therapy Association

American Society of Consultant Pharmacists

AMRPA Board of Directors

Association for Home & Hospice Care of North Carolina/South Carolina Home Care & Hospice Association

Association of Rehabilitation Nurses

Botsford Commons Senior Community

CareFirst

Casa de la Luz Hospice Federation of American Hospitals

Genesis Rehab Services

Healthcare Market Resources, Inc

Illinois HomeCare & Hospice Council

Interim HealthCare

Kessler Institute for Rehabilitation


Kindred Healthcare

LeadingAge

LeadingAge Ohio

Madonna Rehabilitation Hospital – Lincoln

Madonna Rehabilitation Specialty Hospital

McLaren Bay Special Care

National Association for Home Care & Hospice

National Association for the Support of Long Term Care

National Association of Long Term Hospitals

National Hospice and Palliative Care Organization

National Readmission Prevention Collaborative

naviHealth

New Jersey Hospice and Palliative Care Organization

Open Arms Home Health Care

Partners In Home Care

Pennsylvania Homecare Association

Rehabilitation Institute of Chicago

RML Specialty Hospital

Rockburn Institute

Saint Mary’s Home of Erie

Select Medical

Sparrow Health System

The Carolinas Center

Uniform Data System for Medical Rehabilitation

Visiting Nurse Associations of America


2 STAKEHOLDER COMMENTS: GENERAL AND MEASURE-SPECIFIC

This section summarizes the main issues raised by the public comments and provides the measure developer’s responses. Issues raised by at least two commenters are summarized and responded to in this section. All other issues are addressed through responses to individual comments in Appendix A. Each summary comment in this section is assigned a unique identifier to facilitate cross-references with individual comment responses in Appendix A. The same information (verbatim comment and our response) is also presented in table format as a separate attachment to this document, titled “Medicare Spending Per Beneficiary – Post-Acute Care Measures: Public Comment Verbatim Report”.

Subsection 2.1 discusses comments relating to the overall approach and process for creating the MSPB-PAC measures, including the measure intent, development process, and implementation. Subsection 2.2 outlines comments related to elements of episode construction such as opening episodes, defining the episode window, defining treatment and associated services, and excluding clinically unrelated services. Finally, subsection 2.3 discusses the measure calculation: episode-level exclusions, risk adjustment, and the definition of the numerator and denominator.

2.1 Overall Approach and Process

2.1.1 Measure Intent 1) Stakeholder Comment: Five comments expressed general support for the MSPB-PAC

measures.

Response: Thank you for your comments and support. We appreciate your thoughtful feedback and engagement with the development of these MSPB-PAC measures.

2) Stakeholder Comment: 14 comments expressed concern that the MSPB-PAC measures do not provide information about the quality of care delivered by a PAC provider (i.e., they do not assess patient outcomes or quality of services). Some commenters expressed the view that these resource use measures should only be used in conjunction with quality measures.

Response: Thank you for your comments. PAC providers involved in the delivery of high quality care and appropriate discharge planning and post-treatment care coordination would be expected to perform well on these measures since beneficiaries would likely experience fewer costly adverse events. Still, CMS recognizes that resource use measures like the MSPB-PAC measures may not take into account patient outcomes or experience beyond those observable in claims data. Accordingly, CMS recognizes the need to use these MSPB-PAC measures in concert with other quality measures that are designed to capture clinical outcomes of care. As an example, an NQF-endorsed “total estimated Medicare spending per


beneficiary” measure (NQF #2158) for inpatient prospective payment system (IPPS) hospitals is used for the Hospital Value-Based Purchasing (VBP) Program.1 The Hospital VBP is a quality incentive program that evaluates hospital performance based on its Total Performance Score (TPS). The TPS in turn is based on scores in four domains: clinical process of care, patient experience of care, outcome, and efficiency. The hospital MSPB measure is reported under the efficiency domain of the TPS. CMS will incorporate commenters’ feedback on the relationship between resource use and quality use measures into the implementation of the relevant LTCH, SNF, IRF, and HHA QRPs.

3) Stakeholder Comment: Six comments objected to the use of the term “efficiency” when describing the MSPB-PAC measures. Commenters requested that the measures be described as only assessing resource use/utilization, cost, or expense.

Response: Thank you for your comments. We propose to clarify the language for these measures moving forward and in the rulemaking process to be explicit that these are resource use measures that assess a given provider’s Medicare spending as compared to that of other providers in the same PAC setting. As mentioned in Summary Comment ID-2 above, these resource use measures reflect high quality care at a given cost that leads to a lower incidence of adverse outcomes (e.g. hospital readmissions), and therefore lower resource use. Still, CMS recognizes the need to use these MSPB-PAC resource use measures in concert with quality measures to provide a more complete evaluation of a given provider’s performance.

4) Stakeholder Comment: 12 comments supported the four setting-specific MSPB-PAC measures, citing important differences (e.g., regulatory restrictions and patient characteristics) between each setting. Some comments noted that it may be appropriate to reconsider this in the future when data allow cross-setting comparisons.

Response: Thank you for your comments and support.

5) Stakeholder Comment: 11 comments believed that the measures would not achieve their stated goals of supporting public reporting of resource use, providing actionable, transparent information to promote care coordination, and to improve the efficiency of care provided to patients.

Response: Thank you for your comments. The development of the MSPB-PAC measures is mandated by the IMPACT Act, which references the NQF-endorsed hospital MSPB measure. This legislative mandate, along with policy concerns regarding variation in Medicare

1 CMS, “National Provider Call: Hospital Value-Based Purchasing (VBP) Program. Fiscal Year 2016 overview for Beneficiaries, Providers and Stakeholders” (April 29, 2014) http://www.fmqai.com/library/attachment-library/CMS_EDU_Webcast_HVBPFY2016BaselineReport_042914_508.pdf

http://www.fmqai.com/library/attachment-library/CMS_EDU_Webcast_HVBPFY2016BaselineReport_042914_508.pdf


spending in PAC settings, provides context for the development of the MSPB-PAC measures.2 By evaluating a given provider’s risk-adjusted Medicare spending in a defined timeframe as compared to that of the national median provider in the same setting, these resource use measures will allow the recognition of providers that deliver high quality care at lower cost to Medicare, when used in conjunction with other quality measures. As above, we note the example of the hospital MSPB measure as used in the Hospital VBP Program to provide actionable information to providers. We have worked closely with clinicians and stakeholders with expertise from each PAC setting throughout the measure development process to ensure that these measures will achieve their stated goals.

6) Stakeholder Comment: Six comments expressed concern about the MSPB-PAC measures’ alignment with the goals of the IMPACT Act. In particular, commenters believed that the MSPB-PAC measures do not allow comparisons across PAC settings. Some commenters recommended consistency between settings, the development of patient classification groups, and outlining how standardized assessment data will be incorporated. Two comments believed that the measures’ use of episodes and provider-level reporting are inconsistent with the “Medicare spending per beneficiary” mandate of the IMPACT Act.

Response: Thank you for your comments. The IMPACT Act authorizes the Secretary to develop resource use measures, including total estimated Medicare spending per beneficiary, and to require the reporting of standardized assessment data in PAC settings. The IMPACT Act’s reference to “total estimated Medicare spending per beneficiary” is the NQF-endorsed hospital MSPB measure described above3. The hospital MSPB measure evaluates hospitals’ Medicare spending relative to the Medicare spending for the national median hospital during a hospital MSPB episode. It assesses Medicare Part A and Part B payments for services performed by hospitals and other healthcare providers during a hospital MSPB episode, which is comprised of the periods immediately prior to, during, and following a patient’s hospital stay. The hospital MSPB measure considers both episodes and Medicare spending on a provider level. We believe that our similar approach to developing the MSPB-PAC measures aligns with the IMPACT Act’s mandate to develop resource use measures.

Development of the MSPB-PAC measures has occurred in accordance with the tight statutorily mandated timeline in the IMPACT Act. When the standardized outcome measures and patient classification groups are developed, these measures can be reviewed carefully and if appropriate, updated to incorporate the data available at that time. In the absence of this standardized cross-setting data, it is currently not appropriate to make

2 See for example MedPAC, ‘Report to the Congress: Medicare Payment Policy’ (March 2015) chapters 7-11 3 76 FR 51626


comparisons between settings for the reasons outlined by commenters supporting setting-specific measures.

7) Stakeholder Comment: Three comments believed that the measures are a burden for providers or a duplication of existing measures.

Response: Thank you for your comments. Because Medicare fee-for-service (FFS) claims are already reported to the Medicare program for payment purposes, PAC providers will not be required to report any additional data to CMS for calculation of this measure. Thus, there will be no additional data collection burden from the implementation of this measure. The MSPB-PAC measures are being developed as mandated by the IMPACT Act and we are not aware of any existing NQF- or consensus organization-endorsed resource use measures for PAC settings. We therefore do not believe that these measures duplicate existing information.

8) Stakeholder Comment: 13 comments expressed concern over unintended consequences resulting from the measures, such as encouraging providers to “cherry pick” healthier patients or to stint on care. This would adversely affect access to care for complex or high-need patients.

Response: Thank you for your comments. We have worked closely with clinicians throughout the measure development process to ensure that we consider and, as far as possible, mitigate potential unintended consequences arising from these measures. There are three main ways in which the MSPB-PAC measures reduce the possibility of providers selectively treating healthier patients or stinting on necessary services. First, the proposed MSPB-PAC measures are setting-specific in recognition of the important differences between settings such as those raised by stakeholders above in Summary Comment ID-4. This ensures that the measures provide a fair and meaningful assessment of PAC providers’ relative resource use. For instance, within each PAC setting, certain episodes are divided into categories that are only compared to each other, for example HHA partial episode payment (PEP) and low-utilization payment adjustment (LUPA) episodes are compared only with other PEP and LUPA episodes, respectively, and LTCH Standard and Site Neutral episodes are compared only with other Standard and Site Neutral episodes, respectively. This recognizes important clinical and payment system differences between patients in these subcategories.

Second, clinically unrelated service exclusions ensure that patients with health conditions outside of the control of the provider are not counted towards the attributed provider. If certain services that are clinically unrelated to PAC treatment (e.g., chemotherapy) were instead included in the MSPB-PAC episodes, it might incentivize providers to avoid treating complex and/or high-cost patients.


Third, the risk adjustment model takes into account patient health circumstances that affect resource use but are beyond the influence of a given provider. This risk adjustment approach helps ensure that providers are not discouraged from treating patients with high care needs. For example, beneficiaries’ prior care status is taken into account through the use of mutually exclusive clinical case mix categories in recognition of the different expected Medicare spending of patients coming from an acute inpatient hospitalization, an institutional PAC provider (i.e., IRF, LTCH, or SNF), a HHA provider, or the community. The risk adjustment model also flags the use of hospice services during an MSPB-PAC episode window, ensuring that resource use for beneficiaries with these services are only benchmarked against expected spending reflecting the average resource use for similar beneficiaries.

2.1.2 Measure Development Process 9) Stakeholder Comment: 11 comments expressed concern over the short timeframe available

for public comment. Many commenters also requested further opportunities for public comment, some requesting a comment period of at least 30 days. Some comments expressed concern over the short timeframe available for CMS and the measure developer to consider public comment.

Response: Thank you for your comments. The statutorily mandated deadlines of the IMPACT Act have compressed the timeline for the measure development process and limited the amount of time available for public comment and review. Within these strict timelines, CMS and the measure developer have sought public input through the TEP in-person meeting and follow-up e-mail survey, the NQF Measure Applications Partnership (MAP) public comment period and in-person meeting, and this pre-rulemaking public comment period governed by “A Blueprint for the CMS Measures Management System” 4 (Blueprint). Furthermore, to accommodate concerns about the timeframe available for public comment, the Blueprint public comment period was extended from an initial 2-week period to 3.5 weeks to allow stakeholders additional time to submit feedback. There will be further opportunities for stakeholders to review and comment on the measure specifications during rulemaking. The public comment period for each of the proposed rules will be open for 60 days.

We reviewed comments on a rolling basis as they were received during the public comment period. This section of the document summarizes comments raised by 2 or more commenters and provides our responses. Appendix A of this report contains the verbatim

4 CMS, “A Blueprint for the CMs Measures Management System v11.1” (August 2015) https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/MMS/Downloads/Blueprint111.pdf

https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/MMS/Downloads/Blueprint111.pdf


text of the comments received and provides our response either by referring to the subsection that discusses it or by responding individually. We hope that this approach provides the public with an understanding of the key issues that have been raised, while also assuring commenters that we have given careful consideration to their feedback. We thank you for your engagement in this process and for your thoughtful input.

10) Stakeholder Comment: Two comments were concerned that TEP feedback had not been adequately considered. One comment requested explanations for decisions that differ from TEP and public comment feedback.

Response: Thank you for your comments. This report provides our response to public comments received during the 3.5 week public comment period. The report includes comments noting specific TEP concerns and our responses. We thank the TEP for their valuable input in the development of these MSPB-PAC measures. Throughout the remainder of the measure development process, we have considered the feedback provided by the TEP at the in-person meeting and through the follow-up email survey, and incorporated many of their recommendations. For example, we followed the TEP’s recommendation to allow PEP and LUPA claims to trigger an MSPB-PAC HHA episode and conducted additional empirical analyses to determine the appropriate gap length to use across PAC settings for allowing adjacent episodes to be treated as one. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for details of this analysis on collapsing stays. We have not included any pre-trigger spending in the cost of MSPB-PAC episodes, as recommended by the TEP. Based on feedback from the TEP at the in-person meeting and in the follow-up email, the measures use a 30-day post-treatment period. Our currently proposed risk adjustment model includes the following variables that were suggested by the TEP: age, dialysis, hierarchical conditional categories (HCCs), long-term care indicator, patients admitted from the community, transplant, psychosocial factors, and spinal cord injury and paralysis diagnoses. To address the TEP concerns about low episode volume for particular providers, we have also performed analyses of the measures’ reliability in order to set a case minimum for reporting. The TEP’s feedback regarding services that are clinically unrelated to PAC also helped guide the development of the service-level exclusions.

There have however been other suggestions that we have considered but not adopted based upon further analysis or public comment. For example, the MSPB-PAC measures do not prorate claims that begin during the associated services period of an episode and extend past the end of the associated services period. These claims are counted in their entirety towards the cost of the episode. While we considered the TEP panelists’ input to prorate these claims, we determined that proration can distort costs when claims are paid on a prospective payment system (PPS) system. This is because the PPS system reimburses


providers with a fixed payment, regardless of the actual length of stay (subject to certain adjustments). As such, prorating a fixed payment for a longer stay appears cheaper on a per diem basis than a shorter stay. This distorts the Medicare spending captured by the MSPB-PAC measures by making shorter stays seem more expensive than longer stays, when the actual spending was the same for both. The approach to not prorate for claims beginning during the associated services period is consistent with the NQF-endorsed hospital MSPB measure. We also considered a 180-day lookback period based on TEP recommendations and public comment. We found that there is not a consistent increase in the predictive power of the risk adjustment model by using the longer lookback period. Furthermore, incorporating a 180-day lookback period would require the extension of the analogous lookback period used in the episode-level exclusion for beneficiaries who are not continuously enrolled in Medicare Part A and Part B for the lookback plus the entire episode window. Lengthening the lookback period used in this exclusion would remove a large number of MSPB-PAC episodes from providers’ MSPB-PAC measure calculation. Please see Summary Comment ID-43 below for further discussion of our analysis of a 180-day lookback period.

11) Stakeholder Comment: 13 comments expressed that there was insufficient detail in the measures specifications document to allow stakeholders to provide meaningful feedback. Commenters requested data such as statistical analyses in relation to the clinical case mix categories, risk adjustment model, reliability testing of the measures, rationales for key methodological decisions (e.g., treating closely adjacent claims as one episode based on a 7- or fewer-day gap), and complete lists of clinically unrelated service-level exclusions. Some comments requested the inclusion of a glossary or definitions section, including key terms such as “episode-weighted median” and “long-term care indicator”.

Response: Thank you for your comments. The content provided in the draft measures specifications document aimed to include sufficient detail to allow commenters to provide meaningful feedback without being overwhelmingly voluminous or technically complex. We understand that this is a fine balance, and have received feedback in the public comment period requesting both simplified explanations and greater detail. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for details on particular aspects of the proposed measures. The revised MSPB-PAC measure specifications document that will be posted during the FY and CY 2017 rulemaking process will include the full list of clinically unrelated service-level exclusions.

We propose to include a glossary of key terms in the revised measure specifications document. In response to the specific question about the meaning of episode weighting, an example of an episode-weighted median is the following: if there are 2 PAC providers and


one provider had a measure score of 1.5 and another had one of 0.5, but the first had 4 episodes and the second only 1, then the episode-weighted median would be 1.5 (i.e., 0.5, 1.5, 1.5, 1.5, 1.5). The long-term care indicator used in our risk adjustment model identifies beneficiaries who have been institutionalized for at least 90 days in a given year. The indicator is based on 90-day assessments from the Minimum Data Set (MDS) and is calculated based on CMS’ definition of institutionalized individuals.

2.1.3 Implementation This section outlines comments received in relation to the implementation of the MSPB-

PAC measures. We have provided a general response to all these comments as follows.

Response: Thank you for your comments. CMS acknowledges the concerns that stakeholders have raised regarding the implementation of the MSPB-PAC measures. Finalization of the details of the proposed implementation of the MSPB-PAC measures within each PAC setting’s respective QRP as mandated by the IMPACT Act is currently in progress. These details will be provided in the rulemaking process through the FY 2017 notices of proposed rulemaking (NPRMs) for LTCH, IRF, and SNF, and the CY 2017 NPRM for HHA.

12) Stakeholder Comment: 10 comments recommended extensive testing and refinement of the measures before they are implemented for public reporting or payment modification purposes. Commenters recommended that CMS consider a dry run period of confidential provider reporting, voluntary testing periods, and/or pilot programs. Some commenters expressed concern about the short statutory timeframes prior to implementation.

Response: Please see our response above at Section 2.1.3.

13) Stakeholder Comment: Nine comments expressed concern over how the measure will be communicated to the public. Commenters were concerned that the public may interpret the MSPB-PAC measures to be a measure of quality or applicable across PAC settings.


14) Stakeholder Comment: Five comments requested clarification on how the measures will operate in tandem with existing Medicare practices, for example in relation to disputed or disallowed claims processing, feedback reports, and alternative payment models. Some stakeholders asked when final claims data would be available for the measure, given Medicare billing and adjudication practices.


15) Stakeholder Comment: 10 comments recommended that real-time data be made available to providers so that they are able to track beneficiaries’ use of services. Some commenters also


requested sufficient data to enable them to replicate the measure calculation to assess their ongoing performance.


16) Stakeholder Comment: Five comments recommended that there be an appeals or dispute resolution process in the event that a provider disagrees with a given score.


2.2 Episode Construction

2.2.1 Opening (Triggering) Episodes 17) Stakeholder Comment: Five comments requested statistical analyses on the rationale for

collapsing closely adjacent stays based on a gap length of 7 or fewer days. Commenters suggested alternative lengths between stays, such as 9 days under the LTCH interrupted stays policy or IRF’s two-midnight rule. Some commenters also expressed concern about unintended consequences arising from this rule, for example by incentivizing providers to “manage the length of stay” which could adversely affect patient care.

Response: Thank you for your comments. The methodology for collapsing closely adjacent stays for the same beneficiary and provider into one episode was developed to more closely reflect the beneficiary’s trajectory of care with a given PAC provider, and to reduce incentives for providers to influence or control beneficiaries’ length of stay, not for clinical reasons but to improve their performance on the measures. For example, permitting each admission as triggering a new episode might encourage providers to discharge and readmit patients to artificially shorten the length of an episode and reduce the amount of Medicare spending captured. This potential incentive for providers to discharge for non-clinical reasons could adversely affect the continuity of care of a patient and other health outcomes. Treating closely adjacent stays as part of one episode reduces these potential incentives to influence a beneficiary’s length of stay for reasons unrelated to that patient’s clinical status and medical needs.

The 7-day threshold was established by performing empirical analyses of Medicare claims data for each PAC setting. We examined the distribution of gap lengths between adjacent episodes in each setting, and tested different gap lengths and the number of resultant episodes. Based upon feedback from TEP panelists and feedback from CMS clinicians, we used the results of these analyses to inform the determination of a consistent gap length that was common across all PAC settings. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for further details of these analyses.


Regarding the specific gap lengths suggested above, we do not believe that either are an appropriate gap length. The two-midnight rule in the IRF setting establishes the benchmark criteria that should be used when determining whether inpatient admission is reasonable and payable under Medicare Part A, so we do not consider this an appropriate gap length.

18) Stakeholder Comment: Two comments supported treating readmissions to any facility in the same setting as part of the same stay, rather than requiring the beneficiary to be readmitted to same provider.

Response: Thank you for your comments. Readmissions to a different PAC provider of the same type are currently attributed to the first provider’s episode if they occur during the episode window. The readmission by the second provider also triggers its own episode. We do not collapse closely adjacent stays for different PAC providers because then (a) the length of the first provider’s treatment period becomes dependent on a separate provider and (b) the readmission would not trigger its own episode and therefore would not count toward the second provider’s MSPB-PAC measure. As a result, the second provider would not be accountable for the Medicare spending related to their treatment services and services occurring after discharge. This lack of accountability for the second provider would undermine the MSPB-PAC measures’ goal of evaluating Medicare spending throughout a patient’s care trajectory and ensuring that every provider that cares for a patient in that care trajectory has incentives for high quality and low cost care.

19) Stakeholder Comment: Four comments requested clarification on the treatment of LTCH interrupted stays.

Response: Thank you for your comments. An LTCH interrupted stay is reimbursed by Medicare as one claim under the LTCH PPS. As such, the MSPB-PAC LTCH measure will define the episode trigger as the initial admission to the LTCH. The treatment period will begin at the trigger and end at the beneficiary’s final discharge from the LTCH. The associated services period will begin at the trigger and end 30 days after the end of the treatment period. If the duration of the interruption is 3 or fewer days, inpatient and outpatient services provided to the beneficiary during that time are bundled into the one payment to the LTCH and the LTCH is responsible for paying the provider for those services “under arrangements”. “Under arrangements” means that the LTCH will bill and pay for those services performed in another setting and no separate payment will be made to another provider for those services. This single payment to the LTCH will be counted as treatment services in the MSPB-PAC LTCH measure. If the duration of the interruption is greater than 3 days and fewer than the allowed number of days away from the facility as per the LTCH PPS (acute care hospital: 4-9 days, IRF: 4-27 days; and SNF: 4-45 days), there will be a separate Medicare payment to the provider of services. The Medicare spending for the


provider’s services during the interruption (e.g., inpatient (IP) hospital, IRF, or SNF) will be included as associated services in the MSPB-PAC LTCH measure. The single payment to the LTCH for the interrupted stay will again be counted as treatment services in the MSPB-PAC LTCH measure.

Where the LTCH interrupted stay policy does not apply under the LTCH PPS, the MSPB-PAC LTCH measure will treat closely adjacent stays as one episode, as described in the draft measure specifications document. For example, if a beneficiary is admitted to an LTCH, discharged to home for 5 days, and readmitted to that same LTCH, this will lead to a second payment under the LTCH PPS. The MSPB-PAC LTCH measure, however, will collapse these two stays and treat them as one episode. As is the case in other settings, this decision reflects the likelihood that these two closely adjacent stays are part of the same episode of care.

2.2.2 Defining the Episode Window 20) Stakeholder Comment: Six comments recommended that the HHA treatment period should

end upon discharge, rather than a fixed 60-day period. Some commenters recommended this for HHA LUPA episodes while others believed it should apply to all HHA episodes, as there are a variety of reasons why a patient might be discharged before the end of the 60-day period (e.g., the patient is no longer homebound or has met his or her goals).

Response: Thank you for your comments. The home health PPS is based on home health episodes of care which are fixed 60-day periods. As the HHA provider is responsible for rendering HHA services during that entire 60-day period, regardless of when the last visit actually takes place, we define the treatment period to align with payment policy. The HHA provider may discharge the beneficiary from their care before the end of the 60-day period (including for the reasons suggested above) but will still receive the full payment as long as there is not a home health claim for the beneficiary initiated within that same 60-day period, in which case a PEP adjustment would apply. The definition of the MSPB-PAC HHA treatment period based on the relevant Medicare payment policy aligns with the way in which the other MSPB-PAC measures define their treatment periods. Furthermore, to account for the differences in the payment structure and beneficiaries’ clinical characteristics in HHA LUPA episodes, HHA LUPA episodes are only compared to other HHA LUPA episodes during the calculation of the MSPB-PAC HHA measure.

21) Stakeholder Comment: Seven comments expressed concern about premature IP discharges to PAC settings or short PAC stays resulting from inappropriately early discharge. Some commenters recommended establishing a minimum number of days in the PAC facility before an episode is triggered.


Response: Thank you for your comments. The MSPB-PAC measures complement the hospital MSPB measure to ensure that there are consistent incentives for acute hospital and PAC providers to deliver quality care and improve care coordination throughout a patient’s care trajectory. All the hospital MSPB and MSPB-PAC measures include a period during which post-treatment spending is attributed to the provider; this accountability incentivizes acute and PAC providers to engage in appropriate discharge planning and post-treatment care coordination to minimize the likelihood of costly adverse events, such as avoidable hospitalizations. In this way, acute care hospitals and PAC providers are discouraged from prematurely discharging a beneficiary.

We acknowledge commenters’ concerns about patients who have been inappropriately discharged too early from an acute or PAC setting who may then go on to have a short stay in the attributed provider’s facility. Under the design of our measures, this stay would trigger a new MSPB-PAC episode. However, we are concerned that the suggested threshold minimum number of days in a PAC setting before an episode can be triggered would have adverse effects on patients’ access to and continuity of care. Complex beneficiaries may face challenges in receiving the level of care that they need if the MSPB-PAC measures enable providers to avoid triggering an episode for potentially high-cost patients after admission. Patients may be discharged from the PAC provider’s care for reasons unrelated to that patient’s clinical status and medical needs.

22) Stakeholder Comment: Two comments believed that the associated services period was too short. A commenter suggested that a 45-day associated services period would capture beneficiaries who are discharged home after a 14-day stay in a SNF and readmitted after a further 2 weeks without the need for a qualifying hospital stay. A commenter suggested that a 180-day associated services period would better reflect the post-discharge pathways for LTCH patients.

Response: Thank you for your comments. The 30-day post-treatment period was favored by the TEP panelists as an appropriate length of time during which a PAC provider can be held accountable for Medicare Part A and Part B spending, subject to certain clinically unrelated service exclusions. To clarify, in the example of the SNF discharge suggested above, the associated services period begins at the episode trigger (i.e., admission to the SNF) and ends 30 days after the end of the treatment period (i.e., discharge). This means that the MSPB-PAC SNF measure, as currently proposed would capture the above scenario. While a longer period such as 180 days may reflect the recovery period for an LTCH beneficiary, it would also capture services that may not be influenced by the attributed PAC provider. Also, the 30-day post-treatment period is consistent with the NQF-endorsed hospital MSPB measure and aligns with widely adopted quality measures for readmissions and mortality.


2.2.3 Defining Treatment Services 23) Stakeholder Comment: Five comments recommended additional inclusions into the definition

of treatment services. Some comments believed that Part D medication should be included in the MSPB-PAC measures. Some comments believed that all physician-ordered services should be included in the measures. One comment recommended that the measures should include any PAC spending prior to admission to reflect a beneficiary’s continuum of care.

Response: Thank you for your comments. Medicare prescription drug coverage is offered to beneficiaries who must choose to enroll in a Part D plan. As not all beneficiaries elect Part D coverage, including Medicare spending for Part D medication into the MSPB-PAC measures would result in higher episode costs for providers with Part D beneficiaries. This could potentially discourage providers from treating patients with Part D coverage which could have adverse effects on their access to care. As such, we do not believe it is currently appropriate to include Medicare payments made under the Medicare Part D drug payment system.

Regarding the recommendation that the MSPB-PAC measures should include the Medicare spending for all physician-ordered services occurring during the episode window, we are concerned that this might discourage providers from treating patients who have high care needs. Inclusion of services that cannot be reasonably managed by the PAC provider could create incentives for providers to avoid treating patients with certain conditions or complex care needs (e.g., patients requiring chemotherapy) that cannot be fully accounted for in risk adjustment models. As such, we exclude certain clinically unrelated services from the measure over which providers have little or no influence.

Regarding the recommendation that the MSPB-PAC measures should include a beneficiary’s Medicare spending for any PAC services prior to triggering an episode, we are concerned that this might discourage providers from treating patients who are admitted from a high-cost setting or stay (e.g., a patient with a prior LTCH or IRF stay would have considerably higher Medicare spending compared to a patient with a prior HHA stay). This could have adverse effects on access to care for these patients. As such, we do not believe it is appropriate to include Medicare payments made for PAC services prior to triggering an MSPB-PAC episode.

2.2.4 Defining Associated Services 24) Stakeholder Comment: Four comments believed that providers do not have sufficient control

over the patient in the post-treatment period. Commenters were especially concerned about the lack of control of HHA providers.


Response: Thank you for your comments. Including a post-treatment period in the measure creates a continuum of accountability between providers and may incentivize improvements in post-treatment care planning and coordination. It is important to hold providers accountable for a post-treatment period; the alternative scenario of not including a post-treatment period in the measure would risk creating incentives for providers to stint on care during treatment, knowing that they are not responsible for any adverse outcomes after the beneficiary has left their direct care.

The MSPB-PAC measures account for the degree of provider control over associated services by excluding certain Medicare Part A and Part B services that are clinically unrelated to PAC care. Acumen’s clinicians worked with a group of clinicians with expertise in PAC settings to determine the excluded set of services for each setting. While HHA providers may have a lesser degree of control over associated services when compared to institutional settings, the MSPB-PAC HHA measure only compares HHA providers to other HHA providers.

25) Stakeholder Comment: Seven comments requested clarification on the “double counting” of services and/or episodes. Some requested confirmation that the definition of associated services would only apply as long as the MSPB-PAC measures are separate setting-specific measures.

Response: Thank you for your comments. To clarify this concern, we will first review the way in which the MSPB-PAC measures are calculated. The MSPB-PAC measures assess the resource use of a given PAC provider relative to all other providers of the same type and in the same setting. The numerator of each measure is the average risk-adjusted episode spending across all episodes for a given provider, multiplied by the national average episode spending for all PAC providers in the same setting to convert measures into easy-to-interpret dollar terms. The denominator of each measure is the episode-weighted national median of the numerator for all PAC providers in the same setting. Risk-adjusted episode spending is constructed by dividing observed episode spending by expected episode spending (as predicted by a risk adjustment model), where observed spending is constructed by summing the standardized allowed amounts on Medicare Part A and Part B claims in the episode window.

The episode grouping framework used to construct episodes allows for existence of multiple episodes that can overlap, with individual services provided to a given beneficiary assigned to multiple episodes. This occurrence principally arises due to the overlap of individual providers’ episode windows during which services can be attributed to these different providers. At first impression, one may believe that this is problematic and could result in double counting of costs, but this impression is mistaken. In fact, to incentivize cost


efficient and high quality care from each of the providers involved in a patient’s care, it is crucial that they each be assigned episodes to share joint accountability for some sets of services. Sharing assigned services across episodes also encourages coordination and continuum of care; when one provider renders efficient, high quality services, the other providers whose episodes include those services will benefit. This sharing of services across different measures is also common in certain quality measures; such an example would be a measure of the rate of hospital readmissions and measures of the rate of particular types of these readmissions. Double counting of costs is not a problem as long as resource use calculations for episode groups are performed comparatively across providers with assigned episodes, and benchmarks are formulated for comparable provider peer groups.

For example a patient who receives treatment at a SNF, is discharged, and then after a week begins receiving HHA care. In this case, the HHA claim would count as associated services for the SNF episode, and would also trigger a separate home health episode attributed to the HHA. The Medicare claim associated with the HHA’s treatment enters both the SNF’s resource use measure and the HHA’s resource use measure. Double counting costs in this manner would be problematic if the purpose of the MSPB-PAC measures were to simply sum costs across episodes to calculate total Medicare spending. However, the MSPB-PAC measures, as well as the hospital MSPB measure referred to in the IMPACT Act, are not intended for this purpose. Instead, the purpose is to benchmark the average episode resource use attributed to a provider to the comparably measured resource use of other PAC providers of the same type. The SNF’s average episode spending across its episodes is compared to that of other SNFs treating the same types of patients, and the HHA is similarly benchmarked against other HHAs. This approach yields relative resource use measures for the SNF and the HHA that can be used to evaluate individual providers’ performance relative to their peers.

To incentivize high quality and cost efficient care throughout a patient’s continuum of care, it is crucial for both PAC providers in this example to be attributed episodes with overlapping services. If instead only a SNF episode was triggered and not an HHA episode, the HHA would not have similar incentives to provide the patient cost-efficient care because the patient would not contribute to the HHA’s resource use measure. Moreover, some adverse outcomes for the patient that are observable in claims – e.g., hospital readmissions -- could be the result of poor care by both the SNF and the HHA during that part of their respective episode windows that overlap. The framework currently used to construct the MSPB-PAC measures would penalize both the SNF and the HHA for the costs associated with such outcomes, whereas the alternative construction discussed earlier in this paragraph


would penalize only the SNF. Consequently, this episode-grouping framework encourages care coordination and joint accountability, keeping the focus on what is best for the patient.

The IMPACT Act itself implicitly recognizes this important point. MSPB-PAC episodes often begin within 30 days of discharge from an IPPS hospital, and the Hospital VBP Program already includes the NQF-endorsed hospital MSPB measure. Therefore, there are guaranteed to be overlaps between hospital episodes for the HVBP program and MSPB-PAC episodes mandated by the IMPACT Act. Various claims in the overlapping period will be counted once in a hospital episode and counted again in an MSPB-PAC episode. This is desirable for the same reason as in the SNF and HHA example above: the approach ensures that both the hospital and the PAC provider have incentives for high quality and cost efficient care, thereby keeping the focus on improving the patient’s experience through the entire continuum of care.

Similar considerations apply even where a SNF, for example, discharges a patient and then re-admits the patient more than one week later. In this case, the MSPB-PAC framework creates two SNF episodes for this provider, the second overlapping the first. If the second episode were excluded from entering the resource use measure, then the SNF provider would not have the same incentives to provide high quality and cost efficient care for the patient during the second SNF episode stay that it has during the first stay. Moreover, the benchmarking discussed above ensures that the provider is not penalized twice for the same cost. For example, if the first and second episode are the same level of efficiency (the ratio of observed spending to expected spending calculated as discussed above), the MSPB-PAC score based on the two episodes would be identical to the MSPB-PAC score based on just the first episode.

There is an additional practical reason to allow episodes to overlap. Imposing a “clean period” such that a MSPB-PAC episode for one provider can never begin during a MSPB-PAC episode for another provider would require eliminating a large number of MSPB-PAC episodes from the resource use measure calculations, given common care transitions between PAC settings. Requiring a 30-day clean period before triggering an episode would result in the exclusion of a large share of each setting’s episodes. This would not be in line with CMS’ goal of capturing a large share of Medicare PAC spending in the MSPB-PAC resource use measures. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for details of the analysis into the effect of a clean period on the measures.

2.2.5 Excluding Clinically Unrelated Services 26) Stakeholder Comment: Eight comments requested the detailed lists of clinically-determined

service-level exclusions. Commenters noted the difficulty of providing meaningful feedback in the absence of this list. Commenters also requested that the lists be publicly available for


comment before being finalized. One commenter requested details of the impact of these exclusions on the MSPB-PAC measures. Commenters generally expressed support for the examples of clinically unrelated services provided in the draft measures specification document.

Response: Thank you for your comments. We acknowledge the limitations faced by commenters on the draft measures specifications report which outlined our approach in determining service-level exclusions without providing a comprehensive list. The draft measures specifications document aimed to provide sufficient detail to allow meaningful feedback, without being overwhelmingly complex or voluminous. The revised MSPB-PAC measure specifications document that will be posted during the FY 2017 and CY 2017 rulemaking process will include the full list of clinically unrelated service-level exclusions. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for details of the impact of service-level exclusions on provider scores and reliability.

27) Stakeholder Comment: Nine comments recommended that hospice services be excluded from the measure. One comment expressed general concern about the inclusion of hospice services and recommended that if they are to be included, be flagged.

Response: Thank you for your comments. We appreciate commenters’ interest in ensuring that the proposed measures do not disincentive providers from the appropriate use of hospice services. We recognize that beneficiaries receiving hospice services have different characteristics than beneficiaries who are not receiving those services, and that as a result it may be unfair to compare episodes with hospice services to those that do not. At the same time, we believe that including hospice services in the MSPB-PAC measures will promote the delivery of appropriate and efficient hospice care. To accommodate both of these concerns, we propose to include the cost of hospice services in the measures, but to account for the incidence of hospice services by including a dummy (or binary) variable in the risk adjustment models when a hospice claim begins within the beneficiary’s episode window. The dummy variable indicates that the particular MSPB-PAC episode has a hospice claim, which the risk adjustment models then takes into account when predicting expected Medicare spending for that beneficiary. This treatment of hospice services ensures that PAC providers have incentives for the efficient delivery of services, but also accounts for the higher cost of episodes with hospice by comparing them to other episodes with hospice.

28) Stakeholder Comment: Five comments recommended additional service-level exclusions, including where there is a new diagnosis in the post-treatment period, where the beneficiary refuses HHA services. One comment recommended identifying properly associated medical services using the chronic conditions list in use through physician practices, Rural Health Clinics, and Federally Qualified Health Centers.


Response: Thank you for your comments. The MSPB-PAC measures include all Medicare Part A and Part B services occurring during the episode window, with a limited set of service exclusions. MSPB-PAC measures exclude certain services that are clinically unrelated to PAC care so that providers are not discouraged from treating high-cost or medically complex patients. However, the measures are intended to capture services that occur after PAC treatment that may reflect the care delivered by the attributed PAC provider, including services that may correspond to new diagnoses. In general, PAC providers are expected to manage patients’ care needs, including chronic conditions and new acute diagnoses. We note that it may be difficult to accurately identify where a beneficiary refuses a recommended course of treatment. Please see Summary Comment ID-29 for discussion of discharge against medical advice, which is an approximation of identifying a patient who has refused treatment within a particular setting. We encourage stakeholders to review and provide feedback on the complete lists of service-level exclusions during the rulemaking public comment period.

2.2.6 Closing Episodes We did not receive sufficient comments on closing episodes to include any summary

comments in this section.

2.3 Measure Calculation

2.3.1 Implementing Episode-Level Exclusions 29) Stakeholder Comment: Six comments suggested additional episode-level exclusions,

including for patients receiving durable medical equipment, prosthetics, orthotics and supplies (DMEPOS) and end-stage renal disease (ESRD) patients, HHA patients admitted from a setting other than an acute IP hospital, discharge against medical advice, patients who move away from the geographic area in the post-discharge period, and patients who transfer from one short stay hospital to another in the pre-admission period.

Response: Thank you for your comments. ESRD services and a limited number of DMEPOS that are beyond the control or influence of the PAC provider may be excluded if they are clinically unrelated to PAC treatment. The beneficiaries themselves, however, are included in the measure; to exclude these patients would not be in line with CMS’ goal of capturing a large share of Medicare PAC spending in the MSPB-PAC resource use measures.

We acknowledge commenters’ concerns that the MSPB-PAC HHA measure should only cover patients who enter HHA as part of the trajectory from an acute to PAC setting. The figures provided by public comment indicate that approximately only one third of HHA visits are post-acute. To exclude the remaining two-thirds of HHA patients would not be in line with CMS’ goal of capturing a large share of Medicare PAC spending in the MSPB-PAC


resource use measures. We address concerns about the differences between HHA patients who are admitted from an acute IP hospital and those who are admitted from another PAC setting or the community through risk adjustment. Please see Summary Comment ID-36 below for a discussion of our analyses on accounting for the higher acuity of patients with a preceding IP stay.

Excluding patients who are discharged against medical advice may create incentives for providers to use this discharge status code to remove high-cost patients from their MSPB-PAC measure calculation. As such, we do not believe it is appropriate to exclude these patients from the measures.

We acknowledge commenters’ concerns about the degree to which a provider has influence if a beneficiary moves away from the facility’s geographic area. However, we have concerns about the feasibility and reliability of implementing this proposed exclusion using available data that may not be an accurate reflection of a beneficiary’s residence. We also believe that including patients in the measure who move away from a facility’s geographic area does not disproportionately favor or disadvantage any particular providers. We therefore do not believe it is appropriate to exclude these patients from the measures.

The concern about the higher complexity of patients who transfer from one short stay hospital to another in the pre-admission period is accounted for in the risk adjustment model, for example through the HCC indicators and clinical case mix categories.

2.3.2 Risk Adjustment Approach Additional Variables 30) Stakeholder Comment: 17 comments recommended that socioeconomic and/or

sociodemographic factors be included as a variable in the risk adjustment model. Commenters suggested factors such as income, education, insurance status (e.g., dual eligible beneficiaries for Medicare and Medicaid), and race be included as variables in the risk adjustment model.

Response: Thank you for your comments. The NQF is currently undertaking a two-year trial period in which new measures and measures undergoing maintenance review will be assessed to determine if risk adjusting for sociodemographic factors is appropriate. For two years, the NQF will conduct a trial of a temporary policy change that will allow inclusion of sociodemographic factors in the risk-adjustment approach for some performance measures. At the conclusion of the trial, the NQF will determine whether to make this policy change permanent.

Furthermore, the Office of the Assistant Secretary for Planning and Evaluation (ASPE) is conducting research to examine the impact of sociodemographic status on quality measures,


resource use, and other measures under the Medicare program as directed by the IMPACT Act. CMS will closely examine the findings of the ASPE reports and related Secretarial recommendations and consider how they apply to quality programs at such time as they are available. As the MSPB-PAC measures will be submitted for NQF endorsement, we prefer to await the results of this trial and study before deciding whether to risk adjust for socioeconomic and demographic factors. CMS will monitor the results of the trial, studies, and recommendations.

31) Stakeholder Comment: Eight comments recommended that functional status be included as a variable in the risk adjustment models.

Response: Thank you for your comments. CMS recognizes the importance of accounting for beneficiaries’ functional status in the calculation of predicted episode spending. However, given the upcoming introduction of standardized assessment data, CMS prefers that each MSPB-PAC measure’s risk adjustment model not include functional status information derived from the current setting-specific assessment instruments. The inclusion of functional status in the risk adjustment models can be revisited in the future when standardized functional status data are available.

32) Stakeholder Comment: Three comments recommended that behavioral/mental health conditions be included as variables in the risk adjustment models.

Response: Thank you for your comments. Behavioral health conditions are currently included in the proposed risk adjustment model in the form of the following HCC indicators: Drug and Alcohol Psychosis, Drug and Alcohol Dependence, Schizophrenia, and Major Depression, Bipolar, and Paranoid Disorders.

33) Stakeholder Comment: Two comments recommended that home caregiver support be included as a variable in the risk adjustment models.

Response: Thank you for your comments. Information on caregiver support is available in the HHA setting through Outcome and Assessment Information Set (OASIS) assessment data. However, our analyses show that while the caregiver support field is filled out for over 94 percent of linked HHA and OASIS data, over 96 percent of the completed responses have the same value, indicating that caregiver support is available. With little variation in this variable, it is unlikely to improve the predictive power of the MSPB-PAC HHA risk adjustment model. There may also be inherent subjectivity for HHA providers in assigning a value to this field. Accordingly, we do not believe it is appropriate for use in the measure.

34) Stakeholder Comment: Four comments recommended that variables for LTCHs be included in the risk adjustment mode: multiple organ failure, prolonged mechanical ventilator use, and the number of days in an intensive care unit (ICU).


Response: Thank you for your comments. Our proposed inclusion of MS-LTC-DRGs as covariates in the risk adjustment models account for prolonged mechanical ventilator use. In response to public comments, we evaluated whether to adjust for multiple organ failure by examining its impact on the predictive power of the MSPB-PAC LTCH risk adjustment model. We determined that the multiple organ failure dummy variable has a relatively small impact on predicting expected spending (e.g., approximately $600 for LTCH) and is not statistically significant. As the proposed MSPB-PAC risk adjustment models have an interaction term that captures renal failure and congestive heart failure, which are the two most frequent sources of multiple organ failure, we do not believe it is necessary to include multiple organ failure as a new covariate. The length of ICU stay is discussed below under Summary Comment ID-36.

35) Stakeholder Comment: Three comments recommended that factors from a prior acute hospital stay should be included in the risk adjustment models, such as prior emergency room (ER) use, number of hospitalizations, length of stay, and ICU stay.

Response: Thank you for your comments. We discuss the recommendation to adjust for certain factors related to the prior acute hospital stay, including length of prior IP stay and length of ICU stay, below in Summary Comment ID-36. We consider that these variables will adequately account for patient characteristics as reflected in a preceding acute hospital stay.

36) Stakeholder Comment: Two comments recommended that additional variables from other IMPACT Act measures be incorporated into the risk adjustment models.

Response: Thank you for your comments. In response to these comments, we evaluated covariates included in the Potentially Preventable Readmission (PPR) and Discharge to Community (D2C) measures that were not reflected in the MSPB-PAC risk adjustment models. Notably, these measures include risk adjustment variables based on information from the preceding IP stay (as applicable). They account for the length of preceding acute hospital stay based on the following breakdown: 1-7 days, 8-11 days, 12-30 days, and 30+ days.

The PPR and D2C measures also include risk adjustment variables for the length of ICU stays. The PPR variables are based on the following breakdown: 0, 1-2, 3, 4-6, 7-9, 10-13, 14-18, 19-24, and 25+ days, while the D2C has one variable for 1-3 days but is otherwise the same.

We analyzed the effect of including variables for length of prior ICU stay and length of prior IP stay on the predictive ability of the MSPB-PAC risk adjustment models. We used the following breakdown of days for the length of prior ICU stay: 1-2, 3, 4-6, 7-9, 10-13, 14-


18, 19-24, and 25+ days; and for the length of prior IP stay: 8-11, 12-30, and 31+ days. A length of 1-7 days for a prior IP stay is part of the “Prior IP” clinical case mix category, so is omitted from this breakdown. There were sizable and statistically significant coefficients for the variables across the PAC settings, and the predictive power of each model increased. Given these results, we will incorporate these additional variables into the risk adjustment models. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for details on the covariates used in each risk adjustment model.

37) Stakeholder Comment: Three comments recommended that Case-Mix Groups (CMGs) be used for risk adjustment in IRF rather than HCCs.

Response: Thank you for your comments. CMGs are payment category variables used in the IRF setting. Similar to other payment category variables (e.g., MS-LTC-DRGs for the LTCH setting) the CMGs are based in part on the clinical characteristics of a patient. As such, including payment category variables improves the risk adjustment models’ ability to predict spending for an episode by controlling for clinical characteristics. By holding these clinical characteristics as manifested in payment categories constant, the risk adjustment models are then able to focus on other sources of spending variation and better predict expected spending for a beneficiary.

We believe that the use of rehabilitation impairment codes (RICs) may be more appropriate than CMGs as the other covariates incorporated in the risk adjustment model partially account for factors in CMGs (e.g., age and certain HCC indicators). RICs do not account for functional status as CMGs do; functional status information in CMGs is based on the IRF patient assessment instrument (IRF-PAI). CMS prefers that each MSPB-PAC measure’s risk adjustment model not include functional status information derived from the current setting-specific assessment instruments. Applying the same rationale from RICs, we also propose to include MS-LTC-DRGs as covariates in the MSPB-PAC LTCH risk adjustment model. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for details on the RIC and MS-LTC-DRG covariates used in the IRF and LTCH risk adjustment models, respectively. We encourage stakeholders to submit comments on this and other aspects of the proposed measures during the rulemaking process.

Clinical Case Mix Categories

38) Stakeholder Comment: Three comments expressed general support for the clinical case mix categories. Some commented that they appear to have face validity, but requested additional analysis to show how they improve estimated Medicare spending.


Response: Thank you for your comments and support. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for further details on the risk adjustment models.

39) Stakeholder Comment: Four comments expressed concern over small sample sizes for the clinical case mix categories.

Response: Thank you for your comments. Given the changes made to the risk adjustment models as a result of public comment, our analyses show that the predictive power of the risk adjustment model and the reliability of the measure are not substantially improved by running a fully interacted model as compared to using dummy variables for each category. Using dummy variables alone removes the concern about small sample sizes in clinical case mix categories for certain settings. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for further details on the covariates in the risk adjustment models.

40) Stakeholder Comment: Three comments expressed concern about beneficiaries who could fall into several clinical case mix categories. Commenters also requested information about the rationale for the order of priority for clinical case mix categories.

Response: Thank you for your comments. Clinical case mix categories are determined by the most recent claim in the 60 days prior to the start of an MSPB-PAC episode. This means that the vast majority of beneficiaries are assigned to a category using this method. It is infrequent that a beneficiary has multiple claims ending on the same day. It is only for these infrequent instances that the order of priority for clinical case mix categories is used to determine the category in which a beneficiary is classified. The order of priority was developed based on clinical input and empirical analyses. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for further details on the order of priority for clinical case mix categories in the risk adjustment models.

41) Stakeholder Comment: Five comments recommended that the category for “Prior PAC” be divided into each setting, citing concerns that a patient admitted from HHA would be very different from one admitted from an LTCH.

Response: Thank you for your comments. As the comments were primarily concerned with HHA patients being significantly different from those in other PAC settings, we analyzed the effect of including a “Prior PAC - HHA” category. We determined that when we additionally include a dummy variable for "Prior PAC - HHA", the coefficient is statistically different from zero and has a sizeable negative value. This indicates this category is on average less expensive than beneficiaries coming from institutional PAC settings. Given these results, we will include a dummy variable for “Prior PAC - HHA” in the MSPB-PAC


risk adjustment models. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for further details on the risk adjustment models.

Other Risk Adjustment Issues

42) Stakeholder Comment: Six comments recommended that the measures outline how standardized assessment data will be used when available.

Response: Thank you for your comments. The MSPB-PAC measures will be implemented into the SNF, IRF, LTCH, and HHA QRPs according to the timeline mandated by the IMPACT Act. As the IMPACT Act-mandated standardized assessment data is currently in development, CMS will evaluate how to incorporate these data into the measure specifications at that time. Specifically, the inclusion of standardized functional status data in the risk adjustment models can be revisited in the future when these data become available.

43) Stakeholder Comment: Three comments expressed concern over the lookback period. One comment recommended that it be extended from 90 to 180 days. Some comments requested clarification on when the lookback period begins.

Response: Thank you for your comments. To clarify, the lookback period is a set number of days prior to the MSPB-PAC episode trigger. During this lookback period, the MSPB-PAC risk adjustment model examines beneficiaries’ Medicare Part A and Part B claims to gather information that is then used to predict expected episode spending. The concern about the length of the lookback period was originally raised by TEP panelists. In response to their concerns and as raised in public comment, we investigated the impact of moving from a 90- to 180-day lookback period and found that there is not a consistent increase in predictive power using the longer lookback period. Furthermore, incorporating a 180-day lookback period would require the extension of the analogous lookback period used in the episode-level exclusion for beneficiaries who are not continuously enrolled in Medicare Part A and Part B for the lookback plus the entire episode window. Lengthening the lookback period used in this exclusion would remove a large number of MSPB-PAC episodes from providers’ MSPB-PAC measure calculation. As such, we concluded that it is not beneficial to move to a 180-day lookback period. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for further details on our analysis on the lookback period.

44) Stakeholder Comment: Four comments expressed concern over how the model handles high-cost outliers.

Response: Thank you for your comments. The MSPB-PAC measures currently handle high-cost outliers in the step where we exclude episodes with outlier residuals. The residuals for


each episode are calculated as the difference between standardized episode spending and predicted spending.

45) Stakeholder Comment: Four comments expressed concern over the accuracy of claims data in reflecting patient acuity or treatment. Some commenters expressed concern that patients on observation status may in fact be clinically more similar to IP patients, yet would not be placed in a clinical case mix category based on prior IP stay.

Response: Thank you for your comments. In order to reflect PAC providers’ resource use, the MSPB-PAC measures rely on administrative claims data. This is also in line with the NQF-endorsed hospital MSPB measure. Services provided to patients on observation status are shown as outpatient claims such that these patients would not be included in a “Prior IP” clinical case mix category. The clinical case mix categories are only a subset of the full set of risk adjustors in the MSPB-PAC measures; the other covariates in the models can help distinguish patient care needs. The inclusion of functional status data that better reflect patient acuity in the risk adjustment models can be revisited in the future when standardized assessment data are available. Please see the MSPB-PAC Public Comment Summary Report: Supplementary Materials for further details on the risk adjustment models, including the list of covariates used for each measure.

2.3.3 MSPB-PAC Measure Calculation 46) Stakeholder Comment: Eight comments recommended that a geographic-specific (e.g., state

or regional) median should be used instead of the national median, citing differences in cost, patient population, and regulation.

Response: Thank you for your comments. The MSPB-PAC measures account for variation in Medicare spending due to beneficiaries’ health characteristics and geographic differences in Medicare payments through risk adjustment and payment standardization, respectively. Risk adjustment uses patient claims history to account for case-mix variation and other factors that affect resource use but are beyond the influence of the attributed PAC provider. We propose to use the same payment standardization methodology as that used in the NQF-endorsed hospital MSPB measure. This methodology removes geographic payment differences, such as wage index and geographic practice cost index (GPCI), incentive payment adjustments, and other add-on payments that support broader Medicare program goals including indirect graduate medical education (IME) and hospitals serving a disproportionate share of


uninsured patients (DSH).5 Geographic differences in case mix are accounted for in the risk adjustment model.

Given the measures’ use of payment standardization and risk adjustment, calculating PAC provider resource use relative to the national median provider of the same type may also be useful in identifying regional variation and incentivizing providers to reduce this variation, in accordance with the measures’ goals of providing actionable, transparent information to providers.

47) Stakeholder Comment: Four comments recommended that the numerator should use the median, rather than mean. Commenters pointed to the need for consistency between the numerator and denominator and believed that the median would ensure fairness for low volume providers.

Response: Thank you for your comments. The use of the median in the denominator is a normalization that ensures that the episode-weighted median score is 1 and all scores can be interpreted as a percentage of the median. It does not change the rank ordering of providers. Although the average in the numerator is more susceptible to the impact of outliers than a median across a provider’s episodes, it leads to a simpler reliability metric. Also, we reduce the impact of high and low payment outliers on a provider’s score by excluding episodes below the 1st percentile or above the 99th percentile of the residual distribution.

48) Stakeholder Comment: Eight comments believed that the measure calculation is too complex. Some commenters requested simplified explanations suitable for a lay audience.

Response: Thank you for your comments. We propose to include a simplified summary of the MSPB-PAC measures in the revised measure specifications document. First, the MSPB-PAC measure involves the calculation of a ratio for each episode triggered by a PAC provider (the “attributed provider”). Specifically, the measure evaluates the attributed provider’s actual spending on a beneficiary’s episode compared to what they are expected to spend for that episode, given that particular beneficiary’s health characteristics as predicted through the use of a risk adjustment model. An attributed provider with episode spending that is more than expected will yield a ratio with a value greater than 1. For example, if a provider spends $1,000 in treating a beneficiary who is predicted through risk adjustment to require $900 of services in the episode, it would be calculated as $1,000/$900 = 1.1. Conversely, if a provider spends less in treating a beneficiary than they would be expected, this will result in a ratio with a value less than 1. For example, if a provider spends $1,000 in

5 QualityNet, “CMS Price (Payment) Standardization – Detailed Methods” (Revised May 2015) https://qualitynet.org/dcs/ContentServer?c=Page&pagename=QnetPublic%2FPage%2FQnetTier4&cid=1228772057350

https://qualitynet.org/dcs/ContentServer?c=Page&pagename=QnetPublic%2FPage%2FQnetTier4&cid=1228772057350


treating a beneficiary who is predicted through risk adjustment to require $1,200 of services in the episode, it would be calculated as $1,000/$1,200 = 0.8.

Second, the measure calculation takes the average of this across all episodes for the attributed provider during a performance period. Similar to the step above, a value greater than 1 indicates that overall, the provider’s actual Medicare spending was more than expected. A value less than 1 indicates that overall, the provider’s actual Medicare spending was less than expected in that performance period. For example, if a provider treats four beneficiaries and the ratio of actual over expected spending for each beneficiary was 1.1, 0.8. 1.3, and 1.2, the average for the provider over that performance period would be 1.1 indicating that their overall Medicare spending was higher than expected after accounting for the characteristics of the beneficiaries.

Third, the measure calculation compares each provider’s score to other providers nationally in the same PAC setting. This is done by multiplying a provider’s average ratio for beneficiaries treated in that performance period with the average episode spending for all PAC providers in the same setting, nationally. This amount is called the MSPB-PAC Amount for that provider. For example, if the average Medicare spending per episode for a provider’s PAC setting is $5,000, this is multiplied with the value of 1.1 as determined in the preceding step, giving an MSPB-PAC Amount of $5,500.

Finally, a given provider’s MSPB-PAC Amount is then divided by the national median MSPB-PAC Amount for that same PAC setting. For example, if the national median MSPB-PAC Amount is $4,000, this hypothetical provider’s measure would be calculated as $5,500/$4,000 = 1.375. This is the provider’s MSPB-PAC score. As the value is greater than 1, it indicates that the provider’s Medicare spending was higher than the national median Medicare spending for that PAC setting.

49) Stakeholder Comment: Two comments requested clarification on how the MSPB-PAC LTCH measure will be calculated given the separate treatment of standard and site neutral episodes.

Response: Thank you for your comments. The measure calculation is performed separately for MSPB-PAC LTCH standard and site neutral episodes to ensure that they are compared only to other standard and site neutral episodes, respectively, and the final MSPB-PAC LTCH measure will combine the two ratios to construct one provider score.

The MSPB-PAC LTCH measure is calculated as the ratio of the MSPB-PAC Amount for each LTCH provider divided by the episode-weighted median MSPB-PAC Amount across all LTCH providers. To calculate the MSPB-PAC Amount for each LTCH provider, one calculates the ratio of the standardized spending for LTCH standard episodes over the expected spending (as predicted in risk adjustment) for LTCH standard episodes, and the


ratio of the standardized spending for LTCH site neutral episodes over the expected spending (as predicted in risk adjustment) for LTCH site neutral episodes, and then averages these ratios across all episodes for the attributed provider. This quantity is then multiplied by the average episode spending level across all LTCH providers nationally for standard and site neutral episodes. The denominator for an LTCH provider’s MSPB-PAC LTCH measure is the episode-weighted national median of the MSPB-PAC Amounts across all LTCH providers. An MSPB-PAC LTCH measure of less than 1 indicates that a given LTCH provider’s Medicare spending is less than that of the national median LTCH provider during a performance period.


3 OVERALL ANALYSIS AND RECOMMENDATIONS

This section discusses our analysis and recommendations relating to the feedback received during the public comment period. Section 3.1 outlines our preliminary recommendations and Section 3.2 is an overall analysis of the comments and our recommendations.

3.1 Preliminary Recommendations

We are reviewing the commenter suggestions with CMS. We will revise the measure specifications document to include the information requested and be presented in a form that is accessible, including the complete list of service-level exclusions, a glossary of key terms, and a simplified explanation of the measure calculation. We will review the risk adjustment models of the MSPB-PAC measures to analyze whether and how their predictive power can be improved using the recommendations of the public commenters.

3.2 Overall Analysis of the Comments and Recommendations

Feedback received on the MSPB-PAC measures was highly constructive. We appreciate the time and thoughtfulness of all who submitted comments. Many stakeholders expressed concern about the short timeframe for comment and the insufficient level of detail provided, particularly in relation to risk adjustment analyses and service-level exclusions. Many commenters were concerned about the implementation of the measures, particularly expressing the view that they should only be used in conjunction with quality measures. Commenters provided important feedback on the risk adjustment models.


APPENDIX A. PUBLIC COMMENT VERBATIM REPORT

This Appendix A contains an index of the public comments received and the verbatim text of each comment. Section A1 contains an index of the comments received during the public comment period that ran from January 13 to February 5, 2016. Section A2 contains the verbatim comments received during the public comment period and our response to each comment.

A.1 Index of Public Comments Received

This section contains an index of the 45 public comments received during the MSPB-PAC public comment period. Table A1 below is an index of the 45 public comments received that provide feedback on the MSPB-PAC measures. The date is when the email was received in the [email protected] inbox. The email address listed for each comment is the one provided for contact; in some cases the contact person is not the same as the person providing the comment.

We received 12 messages that did not provide feedback on the proposed MSPB-PAC measures, for example messages containing personal medical information, provider-specific comments, and messages sent in error. We consider these comments beyond the scope of the MSPB-PAC measure specifications Blueprint notice and have therefore excluded them from the this public comment summary report.

Table A1. Index of Public Comments Received

No. Measure Set or Measure Date Name, Credentials, and Organization of

Commenter Type of

Organization Contact Email Address

Provided

1 Measure Set 1/25/2016 Richard Chesney, Healthcare Market Resources, Inc. Unknown [email protected]

2 HHA 1/25/2016 Cheryl A. Meyer, President, Illinois HomeCare & Hospice Council

Professional association Not provided

3 Measure Set 1/26/2016 Cheryl Lehman, PhD, RN, CNS-BS, RN-BC, CRRN, President, Association of Rehabilitation Nurses

Professional association [email protected]

4 HHA 1/27/2016 E. Liza Greenberg, RN, MPH. Interim Vice President, Quality and Performance Improvement, Visiting Nurse Associations of America


5 HHA 1/27/2016 Barbara Miller, BSN, Administrator, Open Arms Home Health Care HHA barbmopenarms@mediacomb

b.net

6 HHA 1/27/2016 Vicki Hoak, CEO, Pennsylvania Homecare Association


mailto:[email protected]








Commenter Type of


Provided

7 Measure Set 1/27/2016 Dr. Marc Rothman, Chief Medical Officer, Kindred Healthcare Health system [email protected]

8 SNF 1/27/2016 Joanne M. Wisely, MA, CCC/SLP, VP Legislative Advocacy, Genesis Rehab Services SNF [email protected]

m

9 Measure Set 1/27/2016

1/29/2016 Daniel Ciolek, Associate Vice President, Therapy Advocacy, American Health Care Association


10 SNF 1/27/2016 Bill Carder, Individual Individual [email protected]

11 Measure Set 1/27/2016

Josh Luke, PhD, FACHE, Adjunct Faculty, University of Southern California Sol Price School of Public Policy. Founder, National Readmission Prevention Collaborative

Academic institution,

professional association

[email protected]

12 Measure Set 1/29/2016 Debbie Baer, Corporate Compliance / HIPAA Officer, Risk Manager, Saint Mary’s Home of Erie

Retirement community [email protected]

13 LTCH 1/29/2016 Susan Klanecky, MSN, RN, CCM, CRRN, Vice President of Patient Care, Madonna Rehabilitation Specialty Hospital

LTCH [email protected]

14 Measure Set 1/29/2016 Akinluwa (Akin) Demehin, Senior Associate Director, Policy, American Hospital Association


15 IRF 1/29/2016 Christopher A. Lee, MSPT, FACHE, Vice President Rehabilitation, Madonna Rehabilitation Hospital – Lincoln

IRF [email protected]

16 Measure set 1/29/2016 Cheryl Phillips, MD, Senior VP Public Policy and Advocacy, LeadingAge


17 Measure Set 1/29/2016 Cynthia K. Morton, MPA, Executive Vice President, National Association for the Support of Long Term Care


18 Measure Set 1/29/2016 Sharon L. Dunn, PT, PhD, OCS, President, American Physical Therapy Association

Professional association

[email protected]

19 LTCH 1/29/2016 Cheryl A. Burzynski, MSN, RN, NE-BC, President, McLaren Bay Special Care LTCH [email protected]

g

20 LTCH 1/29/2016 James R. Prister, President & CEO, RML Specialty Hospital LTCH bklikas@rmlspecialtyhospital

.org

21 IRF 1/29/2016 Peggy Kirk, Senior Vice President, Chief Clinical Operating Officer, Rehabilitation Institute of Chicago

IRF [email protected]


















Commenter Type of


Provided

22 IRF 1/29/2016

Bruce M. Gans, MD, Chair, AMRPA Board of Directors; Executive Vice President and Chief Medical Officer, Kessler Institute for Rehabilitation; National Medical Director for Rehabilitation, Select Medical


23 Measure Set 1/29/2016 Jayne Hart Chambers, Senior Vice President Quality, Federation of American Hospitals


24 Measure Set 1/29/2016 Dale N. Schumacher, MD, MPH, President, Rockburn Institute Unknown [email protected]

25 HHA 1/29/2016 Rev. Daniel P. Clark, RN, Individual Individual [email protected]

26 Measure set 1/29/2016 Kate Jones, MSN, RN, CCM, Senior Vice President and Chief Clinical Officer, Amedisys, Inc.

HHA and hospice [email protected]

27 HHA 1/29/2016

2/5/2016

Mary K. Carr, V.P. for Regulatory Affairs, National Association for Home Care & Hospice

Addendum Theresa M. Forster, VP for Hospice Policy & Programs, National Association for Home Care & Hospice


[email protected]

[email protected]

28 HHA 1/29/2016

Heather P. Jones, MPH, CHES, COS-C, Associate Vice President of Quality Initiatives & State Relations, SC, Association for Home & Hospice Care of North Carolina/South Carolina Home Care & Hospice Association


[email protected]

29 Measure Set 1/29/2016 Callene Bentoncoury, Administrator, Casa de la Luz Hospice Hospice [email protected]

30 LTCH 1/29/2016 Cherri Burzynski, MSN, RN, NE-BC, President, National Association of Long Term Hospitals


31 HHA 1/29/2016 Theresa L. Lee, JD, MPH, Executive Director, Alliance for Home Health Quality and Innovation


32 Measure Set 1/29/2016 Scott Laker, MD, Chair, Health Policy & Legislation Committee, American Academy of Physical Medicine and Rehabilitation


33 HHA 1/30/2016 Barbara A. McCann, Chief Industry Officer, Interim HealthCare HHA BarbaraMcCann@InterimHe

althCare.com

34 Measure Set 2/2/2016 Cathy Day, RN, C-OB, MSN, CNML, CJCP, Individual Individual [email protected]

















Commenter Type of


Provided

35 LTCH 2/3/2016 Kira M. Carter-Robertson, MHA, FACHE, VP, Post-Acute Services, Sparrow Health System LTCH [email protected]

36 Measure Set 2/4/2016 R. Claiborne Richards, Jr., Chief Executive Officer naviHealth Unknown [email protected]

37 Measure Set 2/4/2016 Kimberly M. Gimmarro, Executive Assistant/Quality Specialist, Botsford Commons Senior Community

Unknown [email protected]

38 Measure Set 2/5/2016 Mary Ann Starbuck, Chief Executive Officer, CareFirst Unknown [email protected]

39 Measure Set 2/5/2016 J. Donald Schumacher, PsyD, President and CEO, National Hospice and Palliative Care Organization


40 Measure Set 2/5/2016 Troy Hillman, Director, PAC Strategy and Analysis, Uniform Data System for Medical Rehabilitation

PAC data provider [email protected]

41 Measure Set 2/5/2016 Donald L. Pendley, MA, CAE, CFRE, APR, President, New Jersey Hospice and Palliative Care Organization


42 Measure Set 2/5/2016 Frank Grosso, RPh, Executive Director & CEO, American Society of Consultant Pharmacists


43 Measure Set 2/5/2016 Dianne Hansen, MT, MHA, BCHH-C, COS-C, Director of Clinical Operations, Partners In Home Care

Hospice [email protected]

44 Measure Set 2/5/2016 Annette Kiser, Director of Quality & Compliance, The Carolinas Center


45 Measure Set 2/5/2016 Kathryn Brod, President/CEO, LeadingAge Ohio


[email protected]

A.2 Verbatim Comments and Responses

This section contains the verbatim comments received during the public comment period and our response to each comment. This information is also presented in table format in the separate attachment to this public comment summary report titled “Medicare Spending Per Beneficiary – Post-Acute Care Measures: Public Comment Verbatim Report”.













Please note that the verbatim text below has been edited as follows to enhance the readability of this document by a public audience. We have omitted letter or email template details (e.g., date and opening salutation). We have omitted contact details often provided in the concluding paragraph (e.g., “Please do not hesitate to contact me [contact name] at [phone number]”). We have converted footnotes to in-text citations and have retained hyperlinks as provided, including those that do not appear to link correctly. We have noted attachments but have not included them.

COMMENT 1 OF 45 Date: 1/25/2016

Measure Set or Measure: Measure Set

Name, Credentials, Organization, and Email of Commenter: Richard Chesney, Healthcare

Market Resources, Inc. [email protected]

Type of Organization: Unknown

TEXT OF COMMENT:

The proposed methodology for measuring post-acute spending at the provider level does not account for variations in resource utilization and therefore, spending, based on when the organization admits the patient, based on where they are in their recovery process or disease progression. For example, the same joint replacement patient admitted to a home health agency immediately after discharge from a hospital has a different set of needs, and therefore, greater need for therapy visits for a longer period of time than the same patient, who had spent 21 days in a skilled nursing facility prior to their home health admission. Therefore, a legitimate comparison can only be made between providers if the spending is analyzed on all patients admitted at the same point in time. Even that condition has its flaws, since hospitals can behave differently. One hospital may keep a patient in-house longer than a local competitor, who tends to “rush” to discharge patients. Patients from the latter institution are more likely to be re-admitted from the home health agency, which increases the per beneficiary spending of the agency through no fault of its own. It is impractical to say that the home health agency should refuse to admit the patient who is “too early.” Medicare should develop some method of risk adjusting for this situation where hospitals behave differently. Thank you for the opportunity to submit comments on this proposed regulation. RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-21 and 35. COMMENT 2 OF 45 Date: 1/25/2016

Measure Set or Measure: HHA



Name, Credentials, Organization, and Email of Commenter: Cheryl A. Meyer, President,

Illinois HomeCare & Hospice Council

Type of Organization: Professional association

TEXT OF COMMENT:

The Illinois HomeCare & Hospice Council (IHHC) is pleased to have the opportunity to comment on the proposed MSPB-PAC Resource Use Measures released in January, 2016. IHHC represents home care providers and suppliers in Illinois. IHHC members were impressed with the quality of the construction of the proposed measure for the home health sector. It is clear that Acumen, LLC, understands the nuances of the payment system sufficiently to account for them in the development of the measure. While this may seem like a given, IHHC members have not always found this to be the case with all CMS contractors. IHHC is impressed with the episode definition for the home health sector as well as with the definitions of associated services and clinically unrelated services included in the service exclusions. Again, the analysis and decisions show a real understanding of post-acute care, at least in the home health sector. IHHC members were also pleased to see language in the Introduction to the Draft Specifications noting that a measure of post-acute costs alone cannot be used to evaluate the quality of care provided to Medicare beneficiaries, regardless of which post-acute setting is under consideration. The assumption that total post-acute costs should be lower if complications and exacerbations are kept to a minimum in the post-acute phase of care cannot be verified without also examining outcomes and other quality measures. Agreement with this point of view was evident in the introduction. RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-1 and 2. COMMENT 3 OF 45 Date: 1/26/2016


Name, Credentials, Organization, and Email of Commenter: Cheryl Lehman, PhD, RN,

CNS-BS, RN-BC, CRRN, President, Association of Rehabilitation Nurses

[email protected]


TEXT OF COMMENT:

On behalf of the Association of Rehabilitation Nurses (ARN) – representing more than 5,400 rehabilitation nurses and more than 13,000 Certified Registered Rehabilitation Nurses (CRRN) who work to enhance the quality of life for those affected by physical disability and/or chronic



illness - we appreciate the opportunity to comment on the Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) Resource Use Measures. Rehabilitation nursing is a philosophy of care, not a work setting or a phase of treatment. We base our practice on rehabilitative and restorative principles by: (1) managing complex medical issues; (2) collaborating with other specialists; (3) providing ongoing patient/caregiver education; (4) setting goals for maximum independence; and (5) establishing plans of care to maintain optimal wellness. Rehabilitation nurses practice in all settings, including freestanding inpatient rehabilitation facilities (IRFs), hospitals, long-term subacute care facilities/skilled nursing facilities (SNFs), long-term acute care facilities, comprehensive outpatient rehabilitation facilities (CORFs), home health agencies (HHAs), and private practices. Rehabilitation nurses take a holistic approach to meeting patients’ nursing and medical, vocational, educational, environmental, and spiritual needs. We begin to work with individuals and their families soon after the onset of a disabling injury or chronic illness and continue to provide support and care, including patient and family education, which empowers these individuals when they return home, to work, or to school. Rehabilitation nurses also often teach patients and their caregivers how to access systems and resources. ARN supports efforts to ensure people with physical disability and/or chronic illness have access to comprehensive quality care in whichever care setting is most appropriate for them. Specifically, as a part of its mission, ARN stands ready to work with policymakers at the local, state, and federal to advance policies and programs that promote maximum independence for people living with physical disability and/or chronic illness, particularly among the Medicare population. MSPB-PAC SNF Resource Use Measure, MSPB-PAC Home Health Resource Use Measure, MSPB-PAC Long-Term Care Hospital (LTCH) Resource Use Measure, and MSPB-PAC IRF Resource Use Measure ARN appreciates the Centers for Medicare and Medicaid Services (CMS) and Acumen, LLC’s efforts to advance care management and improve the efficiency and coordination of care provided to patients in PAC settings by developing measures that allow for meaningful comparisons between providers in the same PAC setting. The development of quality measures is an important step in the process of increasing the quality of health care and improving patient outcomes. We believe measures should be clinically relevant or representative for a given setting or patient population – measures must be meaningful to be useful. The collection and reporting of measures should not present an undue burden on the organizations or facilities implementing them; there also must be inclusion of patient-reported outcomes and measures that are meaningful to patients, family members, and caregivers. ARN urges CMS and Acumen, LLC to clarify how the information collected by the MSPB-PAC resource use measures will be communicated to patients and providers, as we expect the efficiency of SNFs, LTCHs, HHAs, and IRFs may be difficult to convey to beneficiaries in a meaningful manner. We have concerns the information made available to the public could


unfairly be interpreted as a measurement of the PAC provider’s quality of care, rather than an indicator of the facility’s relative efficiency. We understand, and appreciate, the intent of the MSPB-PAC resource use measures is to ensure patients receive high quality care and address geographic variations in Medicare PAC spending. However, assessing providers’ efficiency based solely on the MSPB-PAC measures is inappropriate, given that some post-acute providers may treat a greater number of medically complex patients who require multi-faceted, highly skilled rehabilitation and treatment than other providers of the same type. ARN believes the MSPB-PAC resource use measures may unintentionally encourage facilities to selectively admit or refuse patients based on the type and complexity of their conditions. It is vitally important that individuals with chronic and disabling conditions are served in a PAC setting that includes the provision of services that will optimize health outcomes and quality of life. The MSPB-PAC resource use measures, which evaluate PAC providers’ efficiency relative to the efficiency of the national median PAC provider of the same type, fail to take into consideration the health needs and desired outcomes of each patient. We are concerned that PAC providers will be assessed based solely on costs per patient, without accounting for the superior patient outcomes facilitated by PAC providers. While ARN supports CMS’s efforts to align the MSPB-PAC measures with the hospital MSPB measure, stipulated by the Improving Medicare Post-Acute Care Transformation (IMPACT) Act, as the development of the MSPB-PAC resource measures continues, we encourage CMS and Acumen, LLC to take into consideration the possible financial incentives for hospitals to prematurely discharge patients to PAC facilities. A recent study, published in the journal Medical Care, suggests that some hospitals may prematurely discharge patients to post-acute settings as a substitute for prolonged inpatient care, thus inflating PAC facilities’ costs, increasing hospital readmission rates, and distorting measurement (Sacks, G. D., Lawson, E. H., Dawes, A. J., Weiss, R. E., Russell, M. M., Brook, R. H., Zingmond, D. S., Ko, C. Y. “Variation in Hospital Use of Postacute Care After Surgery and the Association With Care Quality.” Medical Care 54.2 (2016): 172-179.) Conclusion ARN appreciates the opportunity to provide comments to CMS and Acumen, LLC regarding the MSPB-PAC resource use measures. We are available to work with you, your colleagues, the rehabilitation community, and other stakeholders to develop and implement quality measures that will help to improve the quality of care and outcomes for Medicare beneficiaries with physical disability and/or chronic disease. We thank you for your consideration of our concerns, recommendations and requests. RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 8, 13, and 21. COMMENT 4 OF 45 Date: 1/27/2016



Name, Credentials, Organization, and Email of Commenter: E. Liza Greenberg, RN, MPH.

Interim Vice President, Quality and Performance Improvement, Visiting Nurse Associations of

America [email protected]


TEXT OF COMMENT:

Thank you for the opportunity to comment on the proposed measure, Medicare Spending Per Beneficiary (for Home Health). VNAA is a national membership association that supports, promotes and advocates for mission-driven providers of home health, hospice and palliative care. VNAA’s 130 members are nonprofit home healthcare and hospice agencies from all regions of the country from rural to urban. Our members serve communities in over 33 states, through 600 branches. VNAA has also submitted comments on other IMPACT Act Measures, including Discharge to Community, Potentially Preventable Readmissions, and Drug Regimen Review. We encourage Acumen to review all comments submitted for all IMPACT Act measures, VNAA, and other commenters, as the public comments address many issues applicable to all measures. Key issues to be coordinated across measures and CMS contractors include validity and reliability testing, conversion to ICD-10, and risk adjustment, among others. We particularly encourage CMS and its contractors to develop a unified risk adjustment strategy across all measures (IMPACT and current measures). Thank you for taking our comments into consideration. VNAA’s comments on the MSPB measure are as follows: Medicare Spending Per Beneficiary Measure (MSPB) General Comments:

• We would appreciate if CMS or the contractor could provide information on testing of the measure specifications. The descriptive information provided about the measure is very informative. However, it is difficult to assess whether information provided at the conceptual level can be meaningfully executed with the actual data available, and if the proposed groupings are stable and valid. If the proposed specifications have been applied to existing data sets to obtain valid, reliable results, this information should be provided.

• VNAA remains concerned that an overly broad emphasis on resource use will adversely impact patient care. We strongly believe that home health agencies, PAC providers, and other providers need to focus on delivering the services and supplies needed by patients to support their health and independent living. We hope that CMS will make every effort to ensure resource use measures are always coupled with quality measures, that the risk adjustment methods evolve to ensure highest needs patients receive the services they need, and that patient-centered measurement approaches continue to be developed so that our system continually moves towards understanding and providing care that meets patient goals.

• While VNAA appreciates the challenge of developing measures across settings, we note that the stated intent of IMPACT Act measures is to comparisons across PAC providers. We suggest considering development of risk-adjusted groupings that describe populations of patients, and use them to facilitate comparisons across PAC settings.



• VNAA encourages CMS to evaluate MSPB data carefully as the measure is implemented, and to make adjustments as needed before using it for oversight or payment revisions. Experience in hospital MSPB shows that refinement was necessary after the initial measure was released.

• As with other IMPACT Act measures, we remain highly concerned about the short timeframe for comment, particularly given the complexity of the measure. It is important for CMS to make information on the measures available in a manner that is accessible to providers – the intended audience, and on a timeframe that enables practitioners to understand and develop meaningful comments. Going forward, we encourage CMS to present draft measures to the intended audiences through webinars or conference calls, include more detail on testing and specifications, and translate the more complex statistical concepts to a level appropriate for a practitioner (not statistician) audience.

2.5.2 Denominator • We believe that it is more appropriate to compare the MSPB PAC Amount to a

geographic median of PAC Amounts instead of the national median. There is geographic variation in cost, practice patterns, community resources and populations. Another variable that may impact regional spending is the penetration of Medicare Advantage plan enrollment. In communities with high MA enrollment, the remaining FFS population is likely to be statistically different from the overall population. CMS might consider a phase in from regional to national comparisons, such as is in place for the CJR Demonstration Program.

3.1.2 Episode Window

• We are concerned that a 60 day treatment episode will be applied to the home health episode window, regardless of the amount of time the patient was actually treated. Because of the episode payment system that applies to HHAs only, HHAs are accountable for a longer episode window than other PAC providers; and, for much of the window, the HHA does not have a direct relationship with the patient. Home health agencies are limited in how they interact with the patient after discharge. Under current rules, HHAs can’t see the patients unless they are home bound. Nonetheless, HHAs will continue to be accountable for costs incurred. We also note that some LUPA claims are incurred because the patient chose to discontinue home health services. These patients may be different from patients discharged due to clinical improvements. It is important to protect patient decision-making while not penalizing providers for patients who exercise their choices.

3.1.5 Excluding Clinically Unrelated Services (p. 16) • The contractor appears to be proposing a new method for determining which costs to

exclude, which are different for each PAC setting. Why not use existing models that have been tested? The excluded costs should be aligned with clinical algorithms used in the IMPACT Act PPR measure: e.g. clinical costs that are considered preventable in PPR should be included in MSPB, and those that are not preventable in PPR should be excluded from MSPB.

• The inclusion/exclusion rules should be tested for economic impact (e.g. use an existing data set to test how different includes, exclusions and groupings impact the reported


MSPB results and whether these results are stable across regions and time periods). Results from this testing should be made available and should demonstrate reliability of groupings and settings.

3.2.2 Risk Adjustment Approach p. 20

• We would like to know if the contractor considered using the same risk adjustment methods proposed for other IMPACT Act measures, and if not, why not. Wherever possible we recommend that risk adjustment methods be the same as for other IMPACT Act measures

• Case mix groupings o While the 6 groupings proposed by the contractor appear to have face validity, we

would like to see data or evidence showing that patients in these clusters tend to have similar characteristics and that the clusters are valid in predicting costs.

o Consider testing the case mix groupings for stability with and without significant behavioral health conditions (major depression and other SMI) and adding behavioral health as a co-morbid if it significantly impacts case mix. BH and other social factors may interfere with patient self-management. These are challenges to moving the patient to safe discharge.

o Within the mutually exclusive categories proposed for case mix, there are varying definitions that may need to be considered. For example, patients may be provided with transitional care for observation vs. admitted to ICU. The underlying clinical condition might reflect ICU level impairment, but the claim might only show observation status. CMS will need to develop consistent data management rules to ensure that patients are grouped according to the intensity of services provided, not just by the name of the service (which may vary from facility to facility).

o We support the inclusion of prior PAC stay as a risk adjustment/ grouper factor. We suggest adding a case mix grouper or adjustment capture patients who had acute stay within a window (possibly 30 days) before the direct-from-community HH admission. These patients might have a post discharge community stay and then an admission to HH, and represent a higher acuity overall.

• Other Risk Adjustment o Functional status is a significant predictor of readmission and should be included

as a risk adjustor (Shih, et. al. J Gen Intern Med. 2015 Nov;30(11):1688-95) o Living situation should also be included in the risk adjustment-specifically

whether a patient has an "able and willing" care giver in the home. This specific measure is used in readmission risk stratification tools such as LACE. It would complement functional status for a more accurate indicator of readmission. The both functional status and living situation information can be easily extrapolated from the OASIS assessment.

o We encourage inclusion of patient SES as an adjustor. 3.2.1 Implementing Episode Level Exclusions:

• Please clarify how dual eligible will be addressed. Are they an included or excluded population. If included they should be a separate case mix category.


Thank you for considering VNAA’s comments. RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 6, 9, 11, 12, 13, 20, 30, 31, 32, 33, 36, 41, 45, and 46. COMMENT 5 OF 45 Date: 1/27/2016


Name, Credentials, Organization, and Email of Commenter: Barbara Miller, BSN,

Administrator, Open Arms Home Health Care [email protected]


TEXT OF COMMENT:

There are already measures in place to review appropriate and effective care by home health agencies. We are surveyed for following guidelines and we are also measured by outcomes and patient satisfaction. Why would you allocate more resources and waste the funds to duplicate information regarding home health? There is too much government waste already. RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment ID-7. COMMENT 6 OF 45 Date: 1/27/2016


Name, Credentials, Organization, and Email of Commenter: Vicki Hoak, CEO, Pennsylvania

Homecare Association [email protected]


TEXT OF COMMENT:

I am writing on behalf of the Pennsylvania Homecare Association’s home health member agencies to submit feedback and questions on the draft measure specifications for the Medicare Spending Per Beneficiary – Post Acute Care (MSPB-PAC) Resource Use Measure being developed by Acumen. The standardization of data across post-acute care settings required by the Improving Post-Acute Care Transformation Act of 2014 (IMPACT Act) will enable consumers to make more informed choices when it comes to post-acute care. Our members look forward to the development of a consistent and comparable data set for all post-acute care providers which will allow home health agencies (HHAs) to better demonstrate the important role they play in the care continuum. Below is a summary of the feedback our members have to offer on the draft. General Comments




CMS reports a 5.5 percent increase in Medicare spending in 2014, with per-enrollee spending increasing by 2.4 percent, and so PHA understands the need to examine per beneficiary spending more closely. However, it is important that quality measures like MSPB-PAC do not serve as a disincentive for Medicare providers to connect patients with all necessary support services they need to regain independence following illness or treatment in the interest of preventing Medicare spending. We appreciate that the measure is structured to allow episodes between consecutive PAC providers to overlap, which will promote coordination between hospitals and each successive PAC provider and incentivize cost efficient care throughout the continuum. However, PHA cautions Acumen and CMS that holding PAC providers accountable for the beneficiary’s Medicare spending after discharge could have a chilling effect on providers arranging for the aftercare their patients need for long-term success in the community. The draft specifications seek to measure PAC providers’ Medicare spending, when in fact they are structured to measure the individual patient’s spending. We don’t believe this measure will provide an accurate picture of how well providers can care for patients in a cost effective manner. Rather, it will simply isolate Medicare beneficiaries that are high utilizers, without showing the reasons for that high utilization, be it clinical or otherwise. Many of the comments below assume that the development of the MSPB-PAC measure will eventually lead to associated reimbursement penalties or incentive payments based on providers’ scores. PHA urges CMS to allow full vetting of the measure prior to any associated penalties or incentives, including a test period of at least 12 months where the measure is considered report-only. Associated Services Period PHA providers object to the proposed 30-day associated services period during which HHAs will be accountable for beneficiary spending without the ability to exercise any control over the beneficiary’s actions. One can easily imagine a scenario in which the HHA discharges a patient, the patient sees his community physician two weeks later for a follow up and is prescribed a new medication. Without proper medication instructions from the community physician, the individual could end up in the hospital within the 30-day window through no fault of the HHA. Our members strive to educate patients and families upon discharge about proper dosage and side effects. In fact, Pennsylvania agencies score better than the national average when it comes to improving patients’ ability to correctly administer their own medications (54.3% in PA, 53.2% nationally), but we can only provide education on the list of medications before us at that time. It would be unreasonable to hold the agency responsible for Medicare spending that results from adverse drug interactions involving a drug the patient was prescribed after discharge. In addition, an individual can catch the flu by chance, having nothing to do with proper discharge planning or care instructions. Discharged patients can easily become dehydrated based on their diet choices in the month following home health care. While the HHA can teach the patient how to avoid these illnesses and disease complications, there is no control over his or her actions post-discharge when it comes to communicable diseases or dietary choices.


We appreciate the exclusions for services that are clinically unrelated to PAC treatment such as planned admissions, but PHA suggests the contractor create a mechanism that would exclude Medicare claims that fall outside the scope of the home health episode, such as diagnoses made or medications prescribed after discharge. HHA Episode Collapsing PHA recommends collapsing consecutive HHA episodes into one MSPB-PAC episode to account for the treatment of patients with long-term chronic healthcare needs. We understand this to mean the patient’s associated services period (30 days after the claim is filed) will always fall within another home health episode, and Medicare spending during those 30 days will be double counted. By not collapsing consecutive episodes, agencies who serve chronically ill patients for more than one 60-day episode will consistently see high Medicare spending in the associated services period. The draft specifications do say services will be averaged and not summed across MSPB-PAC episodes, and so PHA requests clarification on how consecutive home health episodes will affect the measure. Categorical Risk Adjustments PHA appreciates the effort to filter measure results according to prior patient trajectory with the six case mix categories for risk assessment. Patients recovering from orthopedic surgery have different needs and risks than patients admitted to home health from other PAC settings. This sorting mechanism will help CMS and agencies better understand how care can be tailored depending on the patient’s path through the healthcare continuum. PHA requests clarification on how patients admitted to home health following inpatient care provided in observation status will be categorized, since Medicare claims data will not show inpatient status. Hospital care might have been provided prior to PAC admission, so these patients should not be considered in the case mix category of community admissions. How will the claims review process correct for observation status? Also missing from the draft specifications is the time period that will be used to determine where a patient falls within the six case mix categories. For instance, a patient might have been discharged to home following a hospital stay for pneumonia and three days later is referred to home health care by the community physician to help control chronic illnesses that led to pneumonia. Does the HHA claim qualify as a community or inpatient admission for purposes of the case mix risk adjustment? PHA suggests using a set period of time as a lookback period to determine, based on diagnoses, whether patient claims should be considered inpatient or community for purposes of proper risk adjustment. To truly understand the ability of effective post-acute care to cut down Medicare beneficiary spending, the measure should omit any claims for dual eligible patients as they cannot accurately be risk adjusted to account for the many unique social and economic needs of these patients. Dual eligibles are more likely than other Medicare beneficiaries to have unmanaged chronic conditions and healthcare needs that are complicated by other economic or social factors. The Consensus Standards Approval Committee of the National Quality Forum (NQF) has recently called for adjusting performance measures for sociodemographic status such as dual eligibility when appropriate. The risk adjustment variables in Appendix C do not include any factors for


socioeconomic status and will not do enough to isolate the work of the PAC provider to trim Medicare spending. The measure should focus on Medicare-only patients until proper risk adjustment for duals is developed. National Median Denominator PHA recommends the use of a state or region-specific median as the denominator used to calculate an HHA’s MSPB-PAC score. This will account for healthcare access factors as well as cost variances across the United States and even within state borders. Pennsylvania, with some large, urban areas with multiple hospitals and health systems and other very rural areas that struggle to access needed healthcare, Medicare costs and cultural diversity of Medicare beneficiaries can vary tremendously. Using one national median MSPB factor will ignore these differences even from state to state. HHA performance is best measured by comparing performance to peers by using a state median or a regional median if possible. Thank you for the opportunity to provide our feedback. RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 5, 8, 12, 24, 28, 30, 40, 43, 44, 45, and 46. Regarding your comment about HHA claims, they are not collapsed into one treatment period because of the significant number of long sequences of consecutive claims (e.g., over 180 days). The length of these sequences of consecutive claims means that patient characteristics and treatment may change significantly over time. Risk adjustment can be more accurately performed using the updated information in each claim within the sequence. In addition, the higher Medicare spending of these episodes composed of long sequences of consecutive claims cannot be fairly compared to the standard 60-day HHA episodes. COMMENT 7 OF 45 Date: 1/27/2016


Name, Credentials, Organization, and Email of Commenter: Dr. Marc Rothman, Chief

Medical Officer, Kindred Healthcare [email protected]

Type of Organization: Health system

TEXT OF COMMENT:

Kindred is pleased to have the opportunity to comment on the proposed set of quality measures related to Medicare Spending Per Beneficiary – Post-Acute Care Resource Use Measure. Kindred Healthcare is the leading provider of post-acute care services, to patients in 2,723 hospitals and post-acute care settings in 47 states. We are focused on delivering post-acute care throughout the full continuum of care, including 95 long-term acute care hospitals, 90 skilled nursing and rehabilitation centers, 18 inpatient rehabilitation hospitals, 101 hospital-based acute rehabilitation units, 626 Kindred at Home health, hospice and non-medical home care sites of service, and with RehabCare as a trusted contract partner in 1,773 unaffiliated sites of service.



Kindred Healthcare’s goal is to provide quality, coordinated care for patients when they need it, in the most appropriate setting through an entire episode. Our national presence and full continuum of integrated care locations provides significant opportunity for people to access the right care and services to support recovery and wellness. Simply put, over the last few years we have dedicated ourselves to building a platform for population health management through our continuum of care from hospital to home. It is widely recognized that post-acute care providers play a significant role in the 90-day episode post-hospital discharge, by transitioning patients home at the highest possible level of function and wellness. Kindred Healthcare has worked diligently to lower healthcare costs by reducing rehospitalizations and lengths of stay in acute care hospitals and throughout an entire episode of care. Kindred endorsed the Improving Medicare Post-Acute Care Transformation (IMPACT) Act of 2014 as an important foundation to pursuing step-wise reforms necessary for value-based post-acute care reforms. Kindred Healthcare supports the development of measures to promote the delivery of high quality care to patients. Kindred agrees with the recognition that the Medicare Spending per Beneficiary (MSPB) measure does “not take into account patient outcomes or experiences” and “must be used in concert with other quality measures.” In this letter, Kindred Healthcare highlights the following comments and concerns on the Medicare Spending per Beneficiary measure for post-acute care laid out within the IMPACT Act. Support Setting-Specific Measure. Kindred supports the current approach of each MSPB measure only comparing like providers with a given PAC setting (i.e., Long-term Acute Care Hospitals (LTCHs) to LTCHs, Inpatient Rehabilitation Facilities (IRFs) to IRFs, etc.). As a provider of services across the entire post-acute care continuum, Kindred understands the different regulatory restrictions unique to each provider setting, such as clinical staffing requirements, which have a direct impact on spending. These site-specific regulatory restrictions make it inappropriate to compare different types of PAC providers at this time. Ensure Sufficient Risk Adjustment. In the list of covariates for risk adjustment there is no mention or use of functional status, and (unlike in the MSPB measures for the acute care hospitals) functional status is a key driver of the plan of care, the mission of the employees, the goals of the families, and often effects the total cost of care, mostly through length of stay considerations in all our settings. Kindred recommends that Acumen consider incorporating functional status into the list of covariates for risk adjustment. Clarification on Risk Adjustment. It is unclear from the draft measure specifications whether hospital or post-acute care claims will be used for risk adjustment. If hospital claims are used, they do not always align to the care delivered in PAC settings. The hospital claim may be inconsistent with the condition for treatment by the PAC provider (e.g., a patient requiring post-discharge recovery for simple pneumonia or requiring ventilator support following an inpatient stay for cardiac surgery). If post-acute care claims are used, additional emphasis needs to be placed on more accurate coding of claims not used to determine payment (e.g., skilled nursing facility claims).


Clarification on Home Health Episode. The home health prospective payment system pays HHAs a predetermined rate for each 60-day episode of home health care. Consistent with payment determinations, the draft measure specifications define a home health treatment period as first day of a HHA claim until 60 days after with an associated services period of 30-days after the treatment period. Home health patients frequently have multiple uninterrupted episodes of home health care (i.e., multiple recertifications). Please clarify how the 30-day post associated services period would apply in instances of multiple recertifications for the home health patient. Clarification on Home Health Exclusion. The draft measure specifications indicate that any home health episode that results from a Request for Anticipated Payment (RAP) will be excluded. Under the current home health payment system, home health agencies submit a RAP for every episode to receive 60 percent of their payment up front and the remainder of the payment when the claim is finalized. It is unclear why RAPs are excluded. Please clarify the rationale for excluding RAPs from the home health episode. Clarification on Skilled Nursing Facility (SNF) Episode. Under the current skilled nursing facility payment system, patients sometime shift from a short (Part A) covered SNF stay to a long (Part B) covered SNF stay. For example, a patient that has exhausted their 100-day SNF Medicare Part A covered stay could continue to receive Part B therapy in the SNF with the rest of their SNF services covered by long-term care insurance, out-of-pocket spending, or Medicaid. Please clarify whether or not these patients would be included in the MSPB calculation. If they are included, please clarify how an episode window would be defined for this patient population. Clarification on Long-Term Care Hospital Measure. With the inception of LTCH criteria, LTCH will soon be paid different rates for LTCH standard cases and site-neutral cases. It was unclear from the draft measure specifications the differences between the measurement of LTCH-standard payment rate cases and site neutral payment rate cases. It was also unclear how the measures would be reported – separately or through a composite score? Please clarify the difference between the LTCH standard and site neutral case MSPB measures and how the different measures will be reported. Concern with Inclusion of Hospice in Associated Services. Kindred is concerned that including hospice in the list of associated services will discourage providers from utilizing hospice services during the episode and delay necessary hospice care until the conclusion of the episode period. Kindred recommends removing hospice from the list of associated services for all PAC provider types. Thank you for the opportunity to provide these comments. RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-1, 2, 4, 27, 31, 45, and 49. Regarding your comment about Medicare Part A, beneficiaries who exhausted their Medicare Part A benefit are not excluded from the measures. As the MSPB-PAC measures are focused on Medicare Parts A and B spending in an episode of skilled nursing care, the treatment period for such a beneficiary would extend to the point at which benefits are exhausted. The episode


window would include the treatment period and 30 days after the treatment period. Including these beneficiaries in the measure ensures that providers have incentives to provide high quality, low cost care to those in need of sustained long-term care. At the same time, this definition of the episode window ensures that the duration of episodes does not extend too far past the end of Part A payments, and therefore does not cause providers to have excessively high resource use for these beneficiaries. This approach is unlikely to result in a scenario where providers are encouraged to select to treat beneficiaries who are close to exhausting their benefit period, as the provider would be financially advantaged only if the private pay or Medicaid payment for the non-Medicare covered component is higher than the Medicare reimbursements. Regarding your comment about RAP claims, they may not trigger MSPB-PAC HHA episodes; only final HHA claims may do so. This results in the exclusion of a small fraction of episodes. By incorporating adequate time for claims to be submitted and processed before claims data are extracted for the measure calculation, this issue is mitigated. While RAP claims may not trigger home health episodes themselves, they are still counted towards MSPB-PAC episodes as associated services if they occur during the episode window of a MSPB-PAC episode. The Hospital MSPB measure similarly counts RAP claims in the post-discharge period of the episode window. Regarding your comment on consecutive HHA claims, the Medicare spending of the second HHA claim would be included in the associated services period of the initial MSPB-PAC HHA episode. The second claim would also trigger its own MSPB-PAC HHA episode. Please see Summary Comment ID-25 for a discussion of overlapping episodes. COMMENT 8 OF 45 Date: 1/27/2016

Measure Set or Measure: SNF

Name, Credentials, Organization, and Email of Commenter: Joanne M. Wisely, MA

CCC/SLP, Vice President-Legislative Advocacy, Genesis Rehab Services/Respiratory Health

Services [email protected]

Type of Organization: SNF

TEXT OF COMMENT:

As a participant of the Technical Expert Panel (TEP) convened to review the proposed Medicare Spending Per Beneficiary - Post Acute Care (MSPB-PAC Resource Use Measure), further comments on this topic are shared in this letter. In addition to the contributions provided during the in-person TEP and post-TEP Survey responses submitted in November 2015, these comments should be considered supplemental to those previously shared. Among the many challenges presented in the current health care environment, accurate and timely measurement of resource utilization presents one of the most promising ways to assure effective oversight and cost control. The opportunity to participate in this process is very much appreciated.



As noted in TEP documents, I am employed by Genesis Rehab Services (GRS). GRS provides rehabilitation services to more than 1700 locations in 45 states and the District of Columbia. The majority of GRS service locations are skilled nursing facilities (SNFs) and we partner with the SNF Provider to assure quality care for all patients placed in these facilities. It is this commitment to comprehensive, high quality care that provokes the concerns identified in the MSPB-PAC Resource Use Measure. As noted during the TEP in-person session and in the supplemental post-meeting survey, there are fundamental challenges with the proposed measure that cannot be ignored. Sparing reiteration of the technical components expressed provided by the American Health Care Association (AHCA) and the National Association for the Support of Long Term Care (NASL), know that I have collaborated with these organizations and I have expressed agreement with the analyses presented. The interpretations and the concerns repeatedly shared are at the most basic level. Only a few examples in need of clarification are noted in this letter but it is important to assure none of the documented explanations provided thus far have quelled the confusion and frustration experienced with this measure. The proposed measure is:

• Poorly constructed • Absent essential content to meet compliance with the IMPACT Act of 2014 • Designed to disregard established CMS process for patient transition of care and

claims processing/denial adjudication

POORLY CONSTRUCTED Consistent with information shared in the Summary of Feedback from the Technical Expert Panel on the Medicare Spending Per Beneficiary -Post-Acute Measures (January 2016) and further explained in the AHCA comments, there appear to be core mathematical concerns with the formula. In the Draft Specifications for the Medicare Spending Per Beneficiary -Post-Acute Care (MSPB-PAC) Resource Use Measures, Provided for Public Comment we find:

• The Risk Adjustment formula (page 7 - public comment document) uses the median in the denominator and the Provider mean in the numerator. This, alone, can easily skew the ratio of a Provider's utilization under certain circumstances.

• On page 8 of the same document, the discussion goes on to explain the use of an "average" in the numerator. When this formula is tested, it appears to create an environment where the beneficiaries who truly need SNF care -- the medically complicated patient requiring extensive care -- could result in undermining the quality status of the SNF. Caring for this type of patient would be considered an "outlier".

These examples are consistent with the questions raised during in-person discussions and are among the many concerns that have yet to be clarified or resolved. ABSENT ESSENTIAL CONTENT The IMPACT Act clearly indicates a requirement to measure Medicare spending per beneficiary. The MSPB-PAC Measure formula and the explanations regarding Provider "efficiency" and "Provider utilization" appear to be in stark contrast to the purpose of the IMPACT Act. If the beneficiary has multiple conditions that contribute to the needed SNF


services, how can a physician ordered service for a verified diagnosis not be "associated" to the essential needs of the patient?

• The terms "efficient" and "efficiency” appear to be used inter-changeably with the terms "utilization" and "resource utilization" through the draft-for-comment document. During discussions, the TEP members discussed the importance of keeping these terms segregated as they are not synonymous. If the recommendation is to change the purpose of this measure to one of "efficiency", then the intent of this entire project is in question. Is it not to meet compliance of the IMPACT Act of 2014 to assure reporting of resource use?

• Consistent with TEP discussions and reiterated in post-TEP survey responses, the term "Associated Services" remains without clear definition. Although these may have clarity in the statistical analysis of a beneficiary with a single diagnosis that is the "trigger" event, the reality is that there are very few Medicare beneficiaries with isolated medical conditions and under Medicare Conditions of Participation, the SNF is required to address all the needs of every beneficiary.

Despite the multiple examples of CMS programs and earlier research reports (see survey responses accompanying this letter) that could be used to include credible and proven medical conditions to include or exclude from the formula, the proposal references "internal" clinicians creating lists of clinically unrelated services. It also omits specifics regarding acquired conditions during the PAC treatment period. Not only is transparency needed to know what exclusions are being identified, but inclusion of credible and long-established CMS program information are essential to assure accurate resource utilization measurement. DISREGARDED CMS PROCESSES Medicare Providers are required to assure clinically appropriate and timely patient care transitions. The system is built on a foundation of service and payment rules that have been ignored and discounted in this proposal.

• Physician ordered services and a certified plan of care are the foundation of the Medicare system.

o Why are internal clinicians creating lists of excluded services that a physician would identify as medically necessary?

• CMS has established programs to identify and properly associated medically necessary services, particularly of those with chronic conditions.

o Why has this proposal not used the chronic conditions list already in use through physician practices, in Rural Health Clinics and in Federally Qualified Health Centers?

• Timely and accurate claim submission for the services provided to any beneficiary is a core requirement of the Medicare Program. This proposal indicates the intent for accuracy and timeliness.

o How will this measure interface with current systems and processes to assure timely claims processing, prompt payment, accurate resource liability and appropriate denied payment adjudication?

o In particular, how will providers be able to evaluate costs and coordinate resource use in real-time?

o If the costs of care are determined at a date much later than when the services delivered, how can timely reporting be achieved?


The most basic requirements of Medicare Providers appear lost in the creativity of this resource utilization measure. Nowhere in this proposal can we find reference or respect of timely, accurate and complete resource liability for the Medicare Provider who provides services in good faith. In a measure that appears to have potential for overlap of measured utilization, formulas that could easily skew accurate representation and disregard for the realities of the CMS timely filing and denied payment adjudication processes, we are left bewildered by what is proposed. There is no question regarding the need for an accurate and timely resource utilization measure. Unfortunately, this measure does not appear to have met many of the most basic needs for accuracy and timely CMS information and/or Provider feedback. As development must continue to evolve towards a viable and useful measure that complies with the intent of the IMPACT Act of 2014, I respectfully request the opportunity to continue participation in this process and I look towards to the work with Acumen and CMS. [attachment: post-TEP survey response] RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-3, 5, 6, 10, 14, 15, 23, 25, 26, 44, and 47. Regarding your general comments, the MSPB-PAC measures provide a timely and accurate depiction of resource use given that they are calculated using Medicare administrative claims data that will be reported to providers at regular intervals. Furthermore, the MSPB-PAC measures provide a complete picture of PAC providers' resource use by assessing all Medicare Part A and B spending during a period that includes PAC treatment as well as 30 days post-treatment. COMMENT 9 OF 45 Date: 1/27/2016 and 1/29/2016


Name, Credentials, Organization, and Email of Commenter: Daniel Ciolek, Associate Vice

President, Therapy Advocacy, American Health Care Association [email protected]


TEXT OF COMMENT:

The American Health Care Association /National Center for Assisted Living (AHCA/NCAL) represents more than 12,000 non-profit and proprietary skilled nursing centers, assisted living communities, sub-acute centers and homes for individuals with intellectual and developmental disabilities. By delivering solutions for quality care, AHCA/NCAL aims to improve the lives of the millions of frail, elderly and individuals with disabilities who receive long term or post-acute care in our member centers each day.

AHCA/NCAL is pleased to have the opportunity to comment on the draft MSPB-PAC measure specifications. The following comments are organized as follows [table of contents]



1. General Comments

Alignment with the IMPACT Act

AHCA/NCAL have been, and remain strong supporters of the principles and objectives of the Improving Medicare Post-Acute Care Transformation (IMPACT) Act of 2014, and are committed to working with you to see that measures required in the law are implemented as intended.

The IMPACT Act established a detailed process through which critically important data and information will be collected, analyzed and synthesized across PAC settings. The thoughtful analysis of these data and appropriate stakeholder engagement in developing meaningful quality and resource use measures could provide the foundation for significant changes to post-acute quality and payment policies aligned with the triple aims of the National Quality Strategy of better care, smarter spending, and healthier people.

You cited in the ‘Introduction’ section of the draft MSPB-PAC measure specifications that Section 2(d)(2)(C) of the IMPACT Act “…stipulates that these measures should align with the hospital MSPB measure in certain ways.” AHCA/NCAL agrees that the statute requires such alignment as appropriate. However, we have some concerns about whether the details of the approach selected are appropriate.

First, Section 2(d)(1)(A) of the IMPACT Act specifies that the MSPB-PAC measure include “total estimated Medicare spending per beneficiary.” It appears instead, that the draft measure specifications have instead created setting-specific total estimated Medicare Spending per Episode (MSPE) measures. Such a construction will not permit meaningful comparisons of spending for beneficiaries with similar characteristics across PAC-Settings. In addition, it will not permit meaningful comparisons of spending for beneficiaries with similar characteristics in the same setting, particularly for beneficiaries with complex needs, as the draft specification rules would instead carve up extended beneficiary stays in home health agencies (HHA), or interrupted stays in Long Term Care Hospitals (LTCH), Inpatient Rehabilitation Facilities (IRF), and Skilled Nursing Facilities (SNF) into multiple MSPB-PAC “episodes” within the same provider.

Evaluating the resources needed to care for beneficiaries with such complex needs is exactly what the IMPACT Act MSPB-PAC measure was intended for. Designing a measure that does not permit a meaningful evaluation of care management within and between PAC providers for this important segment of the population by carving up a beneficiary’s care pathway experience into multiple overlapping “episodes” is counterproductive. It will serve to shield, rather than expose the care needs of the most vulnerable beneficiaries.

AHCA/NCAL recommends that you reconsider options that would permit a more accurate representation of risk-adjusted Medicare spending per beneficiary instead of the current approach that truncates episodes of the most complex need beneficiaries into multiple episodes within and across settings. Second, Section 2(b)(2) of the IMPACT Act requires the alignment of claims data with standardized patient assessment data by October 1, 2018 for LTCH, IRF, and SNF, and by


January 1, 2019 of HHA. Specifically, this section states: “…the Secretary shall match claims data with assessment data pursuant to this section for purposes of assessing prior service use and concurrent service use, such as antecedent hospital or PAC provider use…” It is unclear to AHCA/NCAL exactly how the draft per-episode measure specifications would generate data that could be aligned in a meaningful manner with other beneficiary quality measures to, as you state on page 4, “…provide actionable, transparent information to support PAC providers’ efforts to promote care coordination and improve efficiency of care provided to their patients.”

AHCA/NCAL recommends that you identify and explain how the draft MSPB-PAC measure specifications would permit alignment with the Section 2(b)(2) of the IMPACT requirements, and if not, how you intend to modify the MSPB-PAC measures to permit alignment by FY 2019. Meaningful Stakeholder Input

AHCA/NCAL appreciates the tremendous challenges you face in meeting the statutory timeline constraints of this project. We note that Section 2(a)(2)(E)(ii) of the IMPACT Act stipulates that the application date for the resource use measure is October 1, 2016 for LTCH, IRF, and SNF, and January 1, 2017 for HHA. We also acknowledge that Section 2(e) of the IMPACT act gives the Secretary broad latitude to expedite initial implementation of the measures to meet statutory deadlines.

However, AHCA/NCAL would like to voice our concern that the expedited pace of the MSPB-PAC development process, particularly with respect to the lack of alignment with quality measures, has created a high-risk for unintended consequences related to beneficiary safety and access to care. We note that on an October, 21, 2015 Centers for Medicare and Medicaid Services (CMS) MLN Connects National Provider Call titled The IMPACT Act of 2014 and Data Standardization, CMS officials stated that measure development processes may require up to two years.

In contrast, it appears that the measure development process for the draft MSPB-PAC measure specifications being presented in this document did not begin until sometime in mid-2015, and initial stakeholder exposure to the direction of the measure development through a Technical Expert Panel (TEP) was not until late October 2015. The general public’s first opportunity to review the draft measure specifications began with this January 13, 2016 notice, and we have been only offered a 14-day window to review and provide comment on this complex measure.

AHCA/NCAL would like to share with you the following timeline depicting our memberships’ experience with the development schedule for the MSPB-PAC measure.


1

AHCA/NCAL Member Measure Development Timeline Experience – MSPB-PAC Measure Mid-September 2015 CMS request for technical expert panel (TEP) nominations to

react to the MSPB-PAC measure development work to date October 2, 2015 TEP panelist nomination deadline October 9-15, 2015 TEP panelist acceptance notifications October 27, 2015 Some (not all) TEP panelists receive 175-page TEP packet

(which did not include fully specified measure information) October 29-30, 2015 1.5 day face-to-face TEP meeting November 18, 2015 TEP panelists receive 5+ page questionnaire of 30 questions

requesting feedback on how to develop and specify the measure

November 25, 2015 Due date for TEP panelists to respond to questionnaire November 27, 2015 CMS issues a list of Measures Under Consideration (MUC)

to the Measure Application Partnership (MAP) to comply with Section 1890(a)(2) of the Social Security Act for public comment1 . The list includes 30 measures for IRF, LTCH, SNF, and HH (which again did not include fully specified MSPB-PAC measure information)

December 7, 2015 Due date to submit public comments to National Quality Forum (NQF) for MAP endorsement consideration

December 14-15, 2015

MAP PAC/LTC workgroup in-person meeting to consider the MUC list and public comments

January 13, 2016 CMS Posts MSPB-PAC Draft Specifications on CMS Website for Public Comment (referenced information missing & did not include fully specified measure information)

January 20, 2016 CMS Posts Summary of TEP Feedback Document and TEP participant list to CMS Website (after an AHCA member noted to CMS that it was not attached as specified to January 13, 2016 MSPB-PAC Draft Specifications materials)

January 26- 27, 2016 NQF MAP review and vote on CMS MSBP measure described on MUC list

January 27, 2016 Due date to submit public comments regarding MSPB-PAC Draft Specifications

CMS. List of Measures under Consideration for December 1, 2015. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/QualityMeasures/Downloads/2015-Measures-Under-Consideration-List.pdf

Of note is that our AHCA/NCAL member TEP panelists indicated that they requested more detailed follow-up information, which, other than the November 18, 2015 post-TEP survey which contained no new information, did not occur. Regarding the TEP survey, we were surprised that the Summary of MSPB-PAC TEP Feedback document was not released until one week ago, which allowed us only one week to review and comment on as part of this response. We note from the Summary of MSPB-PAC TEP Feedback document that the TEP members raised numerous points of concern, requested additional analyses, and offered specific recommendations that do not appear to have been addressed in the draft MSPB-PAC measure specifications document. It is reasonable that, due to the expedited timeline, this was an unintended oversight. However, we believe that decisions to implement approaches that differ from TEP and public input should be explained, and be supported by data as applicable.

Additionally, the draft MSPB-PAC measure specifications document discusses extensive analyses related to the identification of treatment services, associated services, exclusions, and risk adjustment associated with these variables, yet very little detailed statistical information was

https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/QualityMeasures/Downloads/2015-Measures-Under-Consideration-List.pdf


provided to the TEP or in the draft MSPB-PAC measure specifications document to help us evaluate whether the decisions made in measure development are appropriate. Transparency is essential. We believe that it is unacceptable for measures of such potential significance on care delivery patterns to be developed by “contractor clinicians” and unnamed “independent clinicians with PAC expertise” without public review. These decisions provide the foundation of this measure. The details of these decisions need to be shared for public review and vetting along with an adequate comment period of at least 30 days. It is quite apparent to AHCA/NCAL members that to-date there has been insufficient measure specification information available, and insufficient response timeline available to permit a thoughtful review of the proposed measures, and little to no opportunity for CMS and your organization, as the CMS contracted measure developers to adequately review the submitted stakeholder feedback prior to the submission to the National Quality Forum (NQF) for review, let alone, for potential implementation during FY 2017.

AHCA/NCAL recommends that additional opportunities for public comment be made available after you are able to provide the public complete information, including analyses, related to the MSPB-PAC measure development prior to implementation. “Resource Use” or “Efficiency” measure?

In this draft MSPB-PAC measure specifications document text you use the term “efficiency” sixteen times and “resource use” only fifteen times. In contrast, the term “efficiency” does not appear within any provision of the IMPACT Act language, but the term “resource use” is repeated 31 times in the same statute.

Efficiency measures must link resource use with outcomes, whereas, resource use measures do not. We note that in the first sentence of the Introduction on page 4 of the draft MSPB-PAC measure specifications document states “The Improving Post-Acute Care Transformation Act of 2014 (IMPACT Act) authorizes the Secretary to develop “resource use measures, including total estimated Medicare spending per beneficiary.” Additionally, the second paragraph on that page starts with “The purpose of the MSPB-PAC measures is to support public reporting of resource use in all four PAC provider settings…” The draft measure specifications presented in this document are “resource use” measures NOT “efficiency” measures.

However, the following unambiguous statements in this draft MSPB-PAC measure specifications document can lead a reader to the belief that the MSPB-PAC measures are instead intended by the measure developers as “efficiency measures” rather than “resource use” measures.

Page 5 – “The hospital MSPB measure was originally established by the Affordable Care Act of 2010 and evaluates hospitals’ efficiency relative to the efficiency of the national median hospital during a hospital MSPB episode.” Page 5 - “Similar to the hospital MSPB measure, the MSPB-PAC measures evaluate a given PAC providers’ efficiency relative to the efficiency of the national median PAC provider during an MSPB-PAC episode.”


Page 5 - “For example, the MSPB-PAC measure for SNFs evaluates SNFs’ efficiency relative to the efficiency of the national median SNF during MSPB-SNF episodes.”

Page 7 - The MSPB-PAC measures evaluate PAC providers’ efficiency relative to the efficiency of the national median PAC provider of the same type.

The use of the term “efficiency” directly or implied in the context of a “resource use” measure can have deleterious unintended consequences of impacting beneficiary access to care and quality of care. Without the pairing of a “resource use” measure with meaningful quality measures (also required under the IMPACT Act), a “resource use” measure used in isolation creates an incentive to decrease resource use through avoiding admissions of complex patients or through withholding care.

AHCA/NCAL strongly recommends that the measure specifications only refer to the MSPB-PAC measures as “resource use” measures. Additionally, any and all references to “efficiency” must emphasize the context that “efficiency” is a representation of how a provider can control costs (i.e. resource use) relative to providing quality care (i.e. meaningful outcomes). 2. Detailed comments specific to the draft MSPB-PAC measure specifications document Section 1 – Introduction (p.4-6)

As discussed in the general comments above, AHCA/NCAL strongly believe that setting-specific Medicare spending per episode measures as presented are inconsistent with IMPACT Act. However, if CMS continues to pursue the path of adopting such measures, AHCA/NCAL offers the following comments in the spirit of making the measure as consistent across settings as possible, so that a true site-neutral MSPB-PAC measure as intended by the IMPACT Act can potentially evolve from this foundation.

We believe that the measure construction as presented appears to undermine the states goal in the second paragraph of the introduction on page 4 that “The purpose of the MSPB-PAC measures is to support public reporting of resource use in all four PAC provider settings as well as to provide actionable, transparent information to support PAC providers’ efforts to promote care coordination and improve the efficiency of care provided to their patients.” The proposed approach is extremely complex and does not offer a transparent mechanism for PAC providers to evaluate in real-time many of the risk-factors and other variables used in the measure construction that may impact their decisions. As the number of unknowns increases, a providers risk tolerance decreases, which may impact beneficiary access to care and quality of care. For example, SNFs in markets or areas accepting greater number of IRF and LTCH discharges will be disadvantaged since these patients are sicker patients since they (A) went to IRF or LTCH from the hospital, and (B) could not be discharged home from IRF or LTCH. The method of including these will create an incentive for SNFs to avoid admitting these patients. Without being presented modeling evidence to the contrary, we do not believe these differences can be adjusted for by regression based risk adjustment.

However, if one constructed the measure to be patients discharged to a PAC provider and then look at costs over the next fixed number of days attributable back to that PAC provider, one has


a more uniform measure that allows across provider setting comparisons and does not double count the costs. The hospital MSBP measure, which is aligned with quality measures, is constructed with a fixed number of days creating an incentive for coordination and efficiency. The proposed MSPB-PAC measures are not aligned with quality measures and do not have standardized fixed durations, but instead duplicate costs attributed to different defined episodes involving the same beneficiary, which create incentives reduce resource use without consideration of clinical outcomes. Section 2 – Measure Information (p.7-9)

In the context of four setting-specific measures, AHCA/NCAL would like to offer the following comments:

• The first sentence of Section 2.5 states: “The MSPB-PAC measures evaluate PAC providers’ efficiency relative to the efficiency of the national median PAC provider of the same type.” This sentence should instead state: “The MSPB-PAC measures evaluate PAC providers’ resource use relative to the resource use of the national median PAC provider of the same type.” As we stated in our general comments above, these are resource use measures NOT efficiency measures. Efficiency measures must link resource use with outcomes. Resource measures do not need to link with outcomes. In fact, the IMPACT Act calls for resource measures as stated in the introduction on page 1.

• The statement on page 7 stating “Specifically, the measures assess the cost to Medicare for services performed by the PAC provider and other healthcare providers during an MSPB-PAC episode” is a bit misleading unless one understands the entire measure construct. Since the term MSBP-PAC episode starts with admission to a PAC provider and ends 30 days after discharge from that PAC provider. It should instead say “Specifically, the measures assess the cost to Medicare for services performed by the PAC provider and other healthcare providers during the time a person receives care from the PAC provider and 30 days after the treatment period ends for that PAC provider.”

• In the Brief Description of Measures on page 7, the general formula for risk adjustment makes sense and we appreciate that you are using expected median not expected mean. This is an important consideration as the data will be skewed, and using the median is more appropriate. However, to be consistent, the numerator should also be the median for the provider. Not using the median for the provider can result in one or two outliers skewing a provider’s ratio, particularly with low volume providers.

• The definition of Numerator on page 8 is confusing. It appears that you are comparing an average in the numerator to a median in the denominator. That does not make statistical sense. If the numerator “Amount is the average risk-adjusted episode cost across all episodes for the attributed provider, multiplied by the national average episode spending level for all PAC providers in the same setting”, and there is a variable time window for each episode in the episode construction, then using average in the numerator will create incentives to avoid admitting or caring for really sick patients or outliers.

• On page 8, it would be very helpful if you could provide definitions for the three data points used in the numerator:

1. Provider standardized episode spending (referenced on page 22 in step 2, but the document does not indicate how ‘average” is calculated. Please specify.


2. Provider expected episode spending (referenced on page 23 as comparable with the hospital measure but does not appear to define how it is applied to MSPB-PAC. Please specify. 3. Average standardized episode spending level across all PAC providers of the same type. Please specify.

• On page 8, in the denominator definition, it would be helpful if you could define “episode-weighted national median of the MSPB-PAC Amounts across all PAC providers in the same setting” and explain how it differs from the “average standardized episode spending across all PAC providers of the same type” used in the numerator.

• The page 8-9 definition of episode concerns us. If the MSPB-PAC measures are to be comparable to the hospital measure, then there should be one fixed window of time standardized across each measure. We believe that presently there is inadequate claim information that would permit adequate risk-adjustment to account for patients with complex needs.

Section 3 – Draft MSPB-PAC Measure Specifications (p. 10-25)

In the context of four setting-specific measures, AHCA/NCAL agrees with the general Episode Construction approach outlined in this section; however we wish to provide comments pertaining to specific components in the following remarks.

• We believe the following statement on page 10 is misleading: “The PAC provider that triggers the episode is the provider to whom the episode is attributed for the purpose of calculating the MSPB-PAC measure.” However, if a person is admitted to a second PAC provider, then the person triggers a new episode for the second PAC provider. So the person’s costs are “attributed to two different providers” for a period of time. We recommend the statement to be revised as follows: “The PAC provider that triggers the episode is the provider to whom the episode is attributed for the purpose of calculating the MSPB-PAC measure for that provider (note: a patient can trigger a different episode for another provider upon admission to a second PAC provider [e.g. patient discharged from SNF to HHA] and spending during these overlapping episodes are attributed to both providers).”

• We disagree with the following statement on page 11: “As a beneficiary moves from one provider to the next in his/her care trajectory, every PAC and hospital provider that the beneficiary encounters will have incentives to deliver cost efficient care.” As we stated above in our general comments, the MSPB-PAC measure is not an “efficiency” measure but “resource measure”. Without pairing a resource measure with quality measures, using a resource measure only, creates an incentive to decrease resource use (e.g. avoid or withhold care). You need to make sure that this resource measure is always paired with quality measures otherwise it could have an unintended effect to decrease access to and quality of care received by Medicare beneficiaries.

• While the following statement on page 11 is technically correct: “As such, services are never double counted within a single MSPB-PAC episode”, such services are in reality double counted across PAC provider measures. This is the main reason why these draft MSPB-PAC measure specifications cannot be used as cross-setting measures as envisioned by the IMPACT Act. Additionally, the same beneficiary’s costs are double counted in LTCH, IRF, and SNF if a readmission occurs in the same provider after an 8-


day break (see Table 1 page 12) and for any HHA patient receiving consecutive 60-day episodes (see Section 3.3.1 on page 7-8). While we agree that averaging services across episodes within and between settings so that services are never double counted within a single episode would mitigate some of the risk that a provider may be reluctant to admit, or may be incentivized to stint on care for beneficiaries with complex needs, this does not justify discounting the fact that patient costs are being attributed to multiple overlapping “episodes” under the draft MSPB-PAC measure specifications approach. Averaging does not eliminate the patient access risk from the decision to use four separate MSPB-PAC measures, it just makes lessens the severity of the risk.

• Similar to several earlier comments, the following statement spanning pages 11-12 provides an incorrect reference to “efficiency” when “resource use” is more appropriate: “Rather, the construction of the numerator and denominator is such that the ratio of observed and predicted episode spending are averaged across all of a given provider’s episodes, in order to provide a dollar-denominated measure of cost efficiency.”

Step 1: Opening (Triggering) Episodes (p. 12-13)

• AHCA/NCAL agrees that within the context of four setting-specific measures, the episode trigger should be the patient’s day of admission to the LTCH, IRF, or SNF facility, or the first day of the home health claim that triggered the episode.

• We also agree that the PAC provider that triggered the episode is attributed the episode. • With regards to LTCH, IRF, and SNF settings with proximate readmissions as reflected

in Table 1, and as discussed in Appendix D.1, AHCA/NCAL agree that readmissions for the same patient and provider within 7 or fewer days can be reasonably considered a continuation of an episode, whereas readmission for the same patient and provider can be reasonably considered a new episode. Our support rationale within the context of four separate MSPB-PAC measures is threefold. First, patients with characteristics that commonly require hospital readmissions may have more difficulty in obtaining initial PAC placement. This approach would help mitigate potential access issues. Second, prior analysis has indicated that the larger the gap between admissions, the greater the likelihood that the beneficiary characteristics have changed significantly, nullifying any accuracy of the risk-adjustment factors of the initial admission. Finally, aligning the episode triggers and specific conditions for when a readmission to an LTCH, IRF, or SNF would trigger a new episode is an important step towards the beneficiary-centered MSPB-PAC measure intended by the IMPACT Act.

Step 2: Defining the Episode Window (p.13-15)

• Similar to our comments pertaining to Episode Triggers above, AHCA/NCAL agrees that within the context of four setting-specific measures, the episode window construction should be aligned as much as feasible would be an important step towards the development of a true beneficiary-centered MSPB-PAC measure intended by the IMPACT Act.

• We agree that in the ideal state, the treatment period should begin at the episode admission trigger and end at discharge.


• Additionally, we agree that within the context of four setting-specific measures, the Associated Services period should begin at the episode admission trigger and end 30 days after the treatment period ends.

• We note that the HHA and LTCH site-neutral work-around approaches described in this section would not be necessary if the proposed measure specification was a true cross-setting beneficiary resource use measure.

Step 3: Defining Treatment Services (p. 15)

• AHCA/NCAL appreciates your efforts at defining Treatment Services as those “…either provided directly or reasonably managed by the attributed PAC provider.”

• Additionally, we agree that the specific PAC provider’s PPS claims, Part B claims that are not otherwise bundled into the respective PAC PPS payment, and DMEPOS claims during the treatment period are all appropriate to attribute to the PAC provider for that episode (subject to certain clinically appropriate Part B and DMEPOS exclusions).

• We also appreciate the descriptions of the rules you developed (as described in Appendix B).

• However, we believe that the information presented in the draft specifications does not appear to address several of our concerns and many comments described in the Summary of MSPB-PAC TEP Feedback document in sufficient detail to permit AHCA/NCAL to be able to provide an informed comment on whether we support or oppose the definition of Treatment Services to be attributed to a PAC provider at this time.

Step 4: Defining Associated Services (p.15)

• Similar to Step 3 above, AHCA/NCAL appreciates your efforts at defining Associated Services as those “…non-treatment services that occur within the associated services period of a given episode.”

• However, we believe that the information presented in the draft specifications does not appear to address several of our concerns and many comments described in the Summary of MSPB-PAC TEP Feedback document in sufficient detail to permit AHCA/NCAL to be able to provide an informed comment on whether we support or oppose the definition of Associated Services to be attributed to a PAC provider at this time

Step 5: Excluding Clinically Unrelated Services (p.15-18)

• Similar to Step 3 and 4 above, AHCA/NCAL appreciates your efforts at excluding clinically unrelated services “…because they are clinically unrelated to PAC care and/or Because PAC providers may have limited influence over certain Medicare services delivered by other providers during the episode window.”

• We agree to the general approach presented for the identification of service categories assessed for exclusion in Table 3 on page 16 and the options for excluding services occurring within the episode window reflected in Table 4 on page 18. However, we strongly recommend that you add a new step (4) on page 17 that states “Perform Public Review to Validate Proposed Exclusions/Identify Oversights.” Transparency is essential. These exclusions decisions provide the foundation of this measure to protect beneficiary


access and care quality. The details of these decisions need to be shared for public review and vetting along with an adequate comment period of at least 30 days.

• There is a footnote about planned readmissions on bottom of page 17 but it is unclear how you are using planned readmission costs in the calculation. Please clarify.

• We believe that Table 4 on page 18 is incomplete. There is no reference to excluding hospitalization costs that were from planned hospital admissions although the Appendix A PAC setting tables on pages 26-30 service level exclusion rows indicate that planned admissions are excluded from treatment and associated service windows. This should be included in Table 4.

• We recommend a modification of Table 1 specific conditions to trigger a new PAC episode as well as the exclusions criteria starting on page 19 as it pertains to planned admissions. We believe that that planned readmissions with LOS of <8 days should trigger new admission. Patients with planned hospital admissions, as opposed to patients with unplanned admissions of 7 days or less, will clearly be readmitted to the PAC provider with a different risk profile than the prior PAC admission, and therefore are more clinically similar to patients with unplanned hospital admissions of 8 or more days that this model classifies as a new episode in Table 1. The PAC provider should not be responsible for additional costs incurred after the patient returns from a planned hospital admission unless the risk factors are revised, which is not an option in this model. The most reasonable solution is to close the PAC treatment period at the time of the planned hospital admission, and exclude any associated service period costs related to the planned hospital admission and follow-up. The subsequent PAC admission following the planned hospital admission would be then treated as a new episode with updated risk factors. The approach we propose has two significant advantages to incentivize shortened hospital stays:

1. PAC providers would be more open to initially admit patients with planned hospital admissions as differences in the patients care needs after the planned hospital admission of 7 days or less will not expose them to more risk, and

2. Hospitals would find that PAC providers would be more receptive to readmitting patients after planned hospital admissions before the 8th day of the hospital stay.

• Finally, we believe that the information presented in the draft specifications does not appear to address several of our concerns and many comments described in the Summary of MSPB-PAC TEP Feedback document in sufficient detail to permit AHCA/NCAL to be able to provide an informed comment on whether we support or oppose the process of excluding clinically unrelated services to be attributed to a PAC provider at this time.

Step 6: Closing Episodes (p.18)

• AHCA/NCAL strongly disagrees with the proposed approach on page 18 to include the full payment for all claims that begin within the episode window to be counted towards the episode costs. Our thoughts concur with the TEP sentiments as reflected in this excerpt from page 11 of the Summary of MSPB-PAC TEP Feedback document: “TEP members strongly felt that when a claim started within the episode window but ended


after its close, the payment for that claim should be prorated when calculating the episode spending.”

• The justification and examples you provide in Appendix D.3 to not represent at least the following two PAC-specific scenarios that differ from the hospital measure.

1.PAC provider payment models, particularly SNF and HHA, are structurally different from the hospital model in that patients with more complex and chronic care needs often receive care for an extended period of time. This is a structural disadvantage that is exacerbated by the proposed approach that would potentially attribute costs to their MSPB-PAC episode for up to 59 days after the end of their attributed treatment period, depending, in part, on the downstream provider’s respective payment model and front office billing practices. This approach would be more appropriate if this were a single cross-setting MSPB-PAC measure that followed a patient through the care spectrum and had one discrete end point, as does the hospital model. However, the proposed approach just exacerbates the double counting of costs being attributed to the different PAC providers involved in the care pathway of a single beneficiary subsequent to their acute care discharge. The proposed approach dis-incentivizes access and care delivery for beneficiaries with complex needs. 2. Beneficiary risk characteristics can change dramatically after discharge for reasons beyond the PAC provider’s control. The episode trigger discussion in page 12 and the related proximate stays discussion in Appendix D.1 on page 38 indicate that your empirical analysis suggested that a gap of 8 days from the PAC provider results in a reduced likelihood that the adjacent stays are related. We do not see in Appendix D.3 any consideration of scenarios where a gap exists between the discharges from one PAC provider to admissions to another PAC provider. For example, a beneficiary is discharged from a SNF to home without follow-up PAC services (could be receiving non-PAC follow-up outpatient therapy services). However, at day 29 after SNF discharge the beneficiary’s physician orders HHA benefit services for a chronic condition exacerbation unrelated to the prior SNF stay. We do not believe that the SNF in this example should be attributed for the entire HHA claim payment for downstream PAC services beyond the 30-day post treatment associated services period when there is a significant gap in the initiation of the such services, particularly for a condition unrelated to the SNF stay. If such unrelated downstream costs cannot be excluded from attribution to the SNF episode, then the next most reasonable and fair option would be to prorate the subsequent HHA claim so that the SNF is not attributed costs beyond 30 days from the SNF discharge.

Measure Calculation (p.19-25)

AHCA/NCAL offer the following comments pertaining to the measure calculation and risk-adjustment approach described in pages 19-25 in the context of four setting-specific MSPB-PAC measures.

Exclusions from All MSPB-PAC Measures (p.19-20)


• AHCA/NCAL agrees with the rationale for exclusion (3) on page 19 which excludes “Any episode in which a patient is not enrolled in Medicare FFS Parts A and B for the entirety of the lookback plus episode window, or is enrolled in Part C for any part of the lookback plus episode window.” This makes sense.

• However, we would like to make an observation that there is signification geographic variation in Medicare Advantage (MA) saturation that could result in small samples with larger variance in areas with high MA saturation. Has the risk-adjustment methodology discussed in Section 3.2.2 explored the potential impacts on current risk adjustment as well as whether the model would be stable with growing MA saturation as the rate of episode-level exclusions would be expected to grow?

Risk Adjustment Approach (p.20-22)

• AHCA/NCAL supports the concept presented on page 20 of using “…clinical case mix categories to segment the PAC population into more clinically homogenous groups” to represent care needs prior to entering a PAC episode.

• The six proposed groups appear to have face validity, and are relatively consistent with our experience. However, AHCA has the following specific concerns:

1. There is no data presented to demonstrate that these proposed groups do generate clinically and financially homogenous groups so that we could provide an informed statement of support or opposition to the approach.

2. We recognize that patients could reasonably have prior care needs that fall into more than one of the six proposed groups and that a hierarchical methodology may be necessary to assign a beneficiary episode to only one group. However, the document does not provide sufficient descriptions or data to support the rationale for the priority order for the clinical case-mix categories.

• The following statement on page 20 indicates that the hospital MSPB risk adjustment model has important limitations that may need to be considered for the MSPB-PAC model. “[The hospital model] does not, however, directly account for differences in intensity and type of care received by beneficiaries prior to entering an episode.” Factors such as such as prior emergency room (ER) use, number of prior hospital admissions, hospital length-of-stay (LOS), and intensive care unit (ICU) stay are the strongest predictors of PAC LOS and rehospitalization. In addition, long-term nursing facility residence prior to hospital admission is a marker for frailty and higher resource needs, and therefore should be included in risk adjustment. It appears that these factors are included at least in part in the six proposed categories, but we request that their inclusion as risk factors be indicated more explicitly. Additionally, variables from the CMS potentially preventable rehospitalization measures that are not present need to be included.

• The MSPB-PAC Risk adjustment also needs to take into consideration, functional and cognitive status as well as mental health status (e.g. depression) as they are very strong predictors of PAC utilization and LOS. Since treatment period is a crude measure of LOS, any factors impacting LOS need to be included. Beginning in FY 2017, these measures can be obtained from the PAC assessment instruments, which also are being standardized as part of the IMPACT Act. It’s easy to link claims with PAC assessment


tools, and the IMPACT Act recommends using such data as practical. While we recognize that such data is not currently available at the current stage of MSPB-PAC measure development purposes, we believe it is essential that the draft measure specifications and risk-adjustment model are designed in a way that will accommodate the introduction of such data as it becomes available. Additionally, the measure specifications should include a clear description of the importance of these factors and the pathway that would need to be followed to assure that the measures are updated in a timely manner to include these critically important factors.

• The general rationale for truncating extreme predicted values makes sense and should be done. However, we are not sure if 1% is the correct level. We understand you want to align with the hospital truncation, but we recommend that you first look at the distribution of data and determine if a different level is more appropriate for each PAC setting. We note that on page 22 of the Summary of MSPB-PAC TEP Feedback document, there are similar concerns raised pertaining to the statistical approach of “winsorization” to remove extreme values. One TEP panelist also suggested capping the spending “…at two standard deviations from a regional spending average as is being done in the Comprehensive Care for Joint Replacement (CJR) model.” We agree that this approach also merits consideration. Finally, we would like to reiterate a point we made earlier related to numerator and denominator values, that even after truncation of extreme values, the data remains skewed and we recommend that median and not average values be used.

• The final statistical definition of the measure in Step 7 on pages 24-25 makes sense, though as currently expressed it is likely to make the interpretation of the measure confusing for the general public. We recommend doing two things.

1. Consider subtracting one from the measure value and express as a percentage, so it is directly expressed as the percent difference from the median.

2. Include a plain English example that reads something like: “The PAC MSPB measure is essentially the percent difference in risk adjusted episode costs from the median. For example, if a facility had a MSPB-PAC value of 20%, then, after risk standardization, Facility X was 20% more expensive than the median MSPB-PAC episode cost.”

Appendices

Appendix A – Episode Specifications (p.26-30)

AHCA/NCAL would like to offer the following comments that apply to all setting specific tables A-1 through A-4.

• Trigger Event – In the context of four setting-specific measures, AHCA/NCAL agrees with the definition of the trigger event in these tables.

• Episode Window – In the context of four setting-specific measures, AHCA/NCAL agrees with the definition of the episode window in these tables.

• Treatment Services – AHCA/NCAL believes there is insufficient information provided in these draft measure specifications to permit an informed comment on the adequacy or appropriateness of these items. We would need a more detailed list and preferably analytic results to be able to comfortably support or oppose any item.


• Associated Services – AHCA/NCAL believes there is insufficient information provided in these draft measure specifications to permit an informed comment on the adequacy or appropriateness of these items. We would need a more detailed list and preferably analytic results to be able to comfortably support or oppose any item.

• Service Exclusions – AHCA/NCAL believes there is insufficient information provided in these draft measure specifications to permit an informed comment on the adequacy or appropriateness of these items. We would need a more detailed list and preferably analytic results to be able to comfortably support or oppose any item.

• Episode Exclusions - AHCA/NCAL believes there is insufficient information provided in these draft measure specifications to permit an informed comment on the adequacy or appropriateness of these items. We would need a more detailed list and preferably analytic results to be able to comfortably support or oppose any item.

• Overall Claim Exclusions - In the context of four setting-specific measures, AHCA/NCAL agrees with the definition of the overall claim exclusions in these tables. However, in the interests of transparency and to enable providers to have a resource to be able to make real-time assessments related to whether a service is excluded or not, we would appreciate more detailed setting-specific lists of claim exclusions.

Appendix B – First Day Service Exclusions (p.31-34)

In the context of four setting-specific measures, AHCA/NCAL agrees with the described approach for identifying first day service exclusions and the related tables B-1 through B-3. However, we recommend the specifications include more detail so that providers can clearly identify those day of admission services that should not be attributed to their MSPB-PAC episode.

Appendix C – Risk Adjustment Variables (p.35-37)

In the context of four setting-specific measures, AHCA/NCAL believes the risk adjustment variables listed in Appendix C, tables C-1 through C-5 are woefully inadequate and do not reflect numerous items that AHCA/NCAL has suggested, and that were recommended by the MSPB-TEP members as reflected in their comments described in the Summary of MSPB-PAC TEP Feedback document. Additionally, no regression analysis results were presented in the draft MSPB-PAC measure specifications document to enable stakeholders to evaluate and provide feedback on the predictive value of the variables included in the measure risk-adjustment model. We believe that both of these concerns must be addressed and resolved before a reliable and valid measure could be established.

[Addendum 1/29/2016]

Exclusions

As AHCA/NCAL continues to evaluate the Section 3 – Draft MSPB-PAC Measure

Specifications section of the draft MSPB-PAC measure specifications document (p. 1025) we have additional broader concerns beyond the specific details we previously commented on.


Specifically, we agree the exclusion of certain services from PAC resource use measure risk adjustment calculations is necessary to permit meaningful comparisons between providers treating similar types of patients. In particular, services furnished during the attributed episode window that are clinically unrelated to the reason for PAC care should be excluded as you describe in Step 5 (p.15-18).

However, in your described process it is unclear to us as to whether you have considered the following scenario, and if so, whether you have identified how to mitigate the potential risks involved.

The existing PAC prospective payment system (PPS) models bundle services billed within the respective per-diem (SNF) or per-episode (LTCH, IRF, HHA) payments based upon a list of service inclusions and exclusions specific to that PAC payment model. These are not the same per each PAC setting.

• Have you identified specific services that you are recommending to be excluded that are currently included in the setting-specific PAC PPS payment for one setting but not others?

Such discrepancies could create unbalanced incentives between PAC settings to care for beneficiaries with care needs whose “MSPB-PAC excluded” care costs are included in their setting-specific PPS payment (and therefore MSPB-PAC attribution), but are excluded from MSPB-PAC attribution in other settings.

AHCA/NCAL recommends that you identify such exclusions discrepancies, and where identified, you describe them and articulate how you intend to mitigate the beneficiary access risks that would be present, as all bundled payments (even the theoretically excluded MSPB-PAC services) would be attributed to the provider.

Additionally, we recognize the short development timeline you have available. However, we are concerned that the apparent large volume of exclusions proposed in the draft MSPB-PAC measure specifications reflects that, in your effort to create a regression model to generate an acceptable predictive model for attributing resource use based on beneficiary characteristics, you appear to have overlooked:

1. Identifying the extensive administrative burdens that would be placed on providers to track the sheer volume and complexity of exclusions proposed, and

2. Describing a transparent process where different providers could share information in real-time during care transitions so that they can evaluate a beneficiary’s resource use as part of their care plan design.

Care transitions between providers and care management within a provider requires transparent real-time information. Providers would want to know those costs that would be attributed to them versus those that would not, so they could focus on those costs they can potentially manage more effectively, which is a key purpose of the MSPB-PAC resource use measure.


The apparently extremely lengthy lists of exclusions, and an exclusions process algorithm that can only be managed in the “black-box” of a software program, will only result in increased administrative effort and therefore reduced care delivery time.

AHCA/NCAL recommends that you consider alternative approaches to reduce the complexity of the MSPB-PAC exclusions process to be more transparent to providers, so that real-time care transition and care plan decisions can be made efficiently.

Alignment with the IMPACT Act

In the comments we submitted on January 27 we discussed several concerns our AHCA/NCAL membership raised related to whether the draft MSPB-PAC measure specifications aligned with the IMPACT Act provisions. We would like to add the following to that list of concerns.

Providers and the public require accurate and transparent information to make informed healthcare decisions. A thoughtfully-developed MSPB-PAC resource use measure, combined with other required IMPACT Act quality measures will serve as the foundation for important PAC payment reforms moving towards a more patient-centered payment model. However, we believe that shortcuts cannot be taken in measure development for the sole purpose of meeting statutory deadlines. A resource use measure that does not reflect beneficiary care needs appropriately generates an increasing degree of risk for reduced access and quality of care at each stage of implementation.

We note that Sections 2(f) and 2(g) of the IMPACT Act describe the following timeline requirements regarding the MSPB-PAC resource use measure after they are implemented as required on October 1, 2016 for LTCH, IRF, and SNF and January 1, 2017 for HHA.

1. Providers would begin receiving confidential feedback reports on October 1, 2017 for LTCH, IRF, and SNF and January 1, 2018 for HHA regarding their performance on the MSPB-PAC resource use measure as compared to other providers, on either a quarterly or annual basis (depending on the content detail), and providers would have an opportunity to review so that errors can be corrected.

2. Public reporting of provider performance on the MSPB-PAC resource use measure (along with other quality measures) compared to other providers would begin on October 1, 2018 for LTCH, IRF, and SNF and January 1, 2019 for HHA.

It is unclear from the description of the scope of your project requirements in the Introduction section of the draft MSPB-PAC measure specifications document (p. 4-6) the extent to which you are required to consider the practicality of CMS’ ability to meet the IMPACT Act required timelines for provider feedback reporting beginning in FY 2018, and for public reporting beginning in FY 2019. However AHCA/NCAL members are extremely concerned that these timeline cannot be met given the current incomplete state of these draft MSPB-PAC measure specifications.

We recognize that no measure will be perfect upon initial implementation, and that ongoing refinements will be necessary to assure that the measure reflects what it is intended to do. We would appreciate a better understanding of the types of measure information that would be


attributed to a provider’s MSPB-PAC resource use measure, and how it would also be available in a transparent manner to a provider:

1.In real-time for transitions and care planning decisions, and

2.In the feedback reports to be disseminated beginning FY 2018 to permit corrections of inaccuracies prior to public reporting.

We would also appreciate a better understanding of what benchmarks should be achieved on the MSPB-PAC measure specifications based upon the provider experience and preponderance of corrections required during the confidential feedback process so that erroneous and misleading information is not released through the public reporting process of implementation.

AHCA/NCAL recommends that you describe:

1. What resource use information and verification tools would be made available to providers in real-time and how it would be made available to providers without violating various laws and regulations pertain to conflict-of-interest, anti-trust, privacy, anti-kickback statutes, etc..

2. What resource use information and verification tools would be made available to providers for resource use attributed to them during both the treatment period as well as the associated period of each episode included in the reported MSPB-PAC measure performance feedback reports.

3. What benchmarks should be achieved on the measure specifications based upon the provider experience and preponderance of corrections required during the confidential provider feedback process so that erroneous and misleading information is not released through the public reporting process of implementation.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 15, 16, 25, 31, 32, 35, 38, 40, 42, 47, and 48.

Regarding your comment on a single cross-setting measure, the setting-specific measures account for distinctions between different types of PAC providers, in terms of the beneficiary risk pool, payment policy, and risk adjustment factors for each setting. A single measure across all PAC settings would fail to account for these important differences by assuming a high degree of commonality in patient characteristics, payment policy definitions, and risk adjustment across PAC settings. For instance, LTCHs may treat patients with serious respiratory issues that are not typically treated in HHAs, SNFs, and IRFs. By assuming commonality, a single cross-setting measure would compromise the degree to which it delivers actionable information and a meaningful comparison between PAC providers. In the future, when standardized assessment data are available, CMS may re-visit this and carefully evaluate whether a single cross-setting MSPB-PAC measure is possible. Regarding your comment on Medicare Advantage saturation, we have extensively investigated small cell size issues in our risk adjustment methodology using FFS beneficiaries' claims from FY 2014, and tailored our approach based on our findings. CMS will make adjustments to this risk adjustment methodology in the future in response to changes in the Medicare program (such


as large changes in FFS enrollment), in order to ensure that spending is predicted as accurately and precisely as possible. Regarding your comment on the statement on page 7, thank you for clarifying the intent of that statement; your interpretation is accurate. Regarding your comment on Table 4 on page 18, all services within an episode window are reviewed for exclusion. Along with management of pre-existing conditions, planned hospital admissions are also clinically reviewed for exclusions. Table 4 mainly demonstrates the options that clinicians have towards excluding such occurrences. Therefore if clinicians see a hospital admission that appears to be planned, any of the selections listed in Table 4 are options for exclusion. For purposes of organizing the document, the options are listed in table 4, but the criteria for exclusion are listed in appendix A as you noted. Regarding your comment on planned readmissions, services or admissions that represent planned hospitalization will be excluded from the cost of the entire episode. Certain planned hospitalizations will are excluded through the service-level exclusions developed for these measures. Regarding your comment on truncation, when necessary, we truncate extremely low predicted values and remove high spending outliers in order to improve the measure calculation. As our risk adjustment approach has incorporated feedback from the TEP and public comment period, the distribution of predicted values in each PAC setting has changed since the production of our original documentation. Accordingly, we have revisited the winsorization step and outlier exclusion step to ensure that our percentile cutoffs are appropriate, given the distributions. Currently, for high spending outliers, we eliminate episodes for which the deviation of observed spending from expected spending (as predicted by the risk adjustment model) is especially high. We appreciate the TEP panelist's comment about capping spending at a set number of standard deviations from a regional spending average. However, we believe that predicted payment-standardized spending provides a more appropriate benchmark than a regional spending average, as it can better reveal regional differences in resource use. Regarding your comment on the definition of the numerator, to calculate the numerator or “MSPB-PAC Amount” for each PAC provider, one calculates each episode’s ratio of standardized episode spending over expected episode spending. Then the measure sums the ratios and finds the average by dividing by the number of episodes. The expected episode spending for each episode is found by applying an ordinary least squares (OLS) regression that estimates the relationship between the independent variables (including age, HCCs, disabled/ESRD enrollment status, long term care indicator, and HCC interactions) and standardized episode payments. Finally, the "average standardized episode spending level across all PAC providers of the same type" is the sum of the observed episode spending for all episodes for the particular setting (e.g., SNF) divided by the number of episodes. Regarding your comment on service exclusions, the service-level exclusions do not apply to services that are included in payments made under a consolidated billing system. The service-level exclusions are considered only in relation to services that are not included in a PAC claim's


PPS payment. The commenter may be concerned about the following example. Service A is assumed to be provided under a PPS payment in an LTCH, but is billed and reimbursed separately in a home health agency. If the payment for a Service A claim is excluded from the measure calculation because it is determined to be out of the control of PAC providers, this will reduce resource use for the home health agency but will not affect resource use in the LTCH. We believe this example provides an illustration of the way in which payment policy makes it preferable to have setting-specific MSPB-PAC measures at this time. Setting-specific measures allow the home health agency to be compared only to other home health agencies, and the LTCH to be compared only to other LTCHs. Regarding your comment on fixed episode length, given the differences between PAC settings in terms of patient acuity, services provided, and length of stay, we do not consider it appropriate to use a fixed episode window across all settings. For example, if we used a fixed 30-day treatment window, this would truncate 60-day HHA claims and fail to capture a large proportion of post-discharge services in LTCHs, where the average length of stay is over 25 days. Defining the episode window in terms of a treatment and associated services period allows the specifics of each PAC setting to be taken into account, and better reflects the care and services actually delivered by a provider. To clarify, the NQF-endorsed hospital MSPB measure uses an exactly analogous approach. It is not constructed with a fixed number of days in the episode window. Rather, the episode window covers the length of the IP stay (which may vary) plus a fixed period of 30 days from discharge. The MSPB-PAC measures have been constructed following this course. COMMENT 10 OF 45 Date: 1/27/2016

Measure Set or Measure: SNF

Name, Credentials, Organization, and Email of Commenter: Bill Carder, Individual

[email protected]

Type of Organization: Individual

TEXT OF COMMENT:

My question relates to post-acute service provision in a swing-bed service delivered in a critical access hospital and how the Resource Use Measures relate to such service provision. CMS ICN006951 May 2014 "Swing Bed Services" depicts "Hospitals, as defined in Section 1861 (e) of the Social Security Act, or CAHs with as Medicare provider agreement ... approval to furnish swing bed services, may use their beds as needed to furnish either acute or Skilled Nursing Facility (SNF)-level care." It continues by noting payments are exempt from SNF PPS payments and are based on 101% of reasonable cost. The draft Measures seem silent in regard to how swing bed post-acute service cost would relate to your proposed measures. I suggest some consideration be given so as to not be comparing SNF PPS payment costs with CAH Swing bed cost as combining the two different payment



systems will likely deliver unintended analytical consequences which may be adverse to the continued financial viability of SNF services provided by a Critical Access Hospital.

I realize the scope of your quest is a focus on “resource use measures, including total estimated Medicare spending per beneficiary” and to require the reporting of standardized assessment data in post-acute care(PAC) settings: skilled nursing facilities (SNFs), home health agencies (HHAs), long-term care hospitals (LTCHs), and inpatient rehabilitation facilities (IRFs).

The swing bed program of a Critical Access Hospital by CMS definition is the provision of SNF level service but it is not a SNF. Accordingly the measures are moot in its application to Critical Access Hospital swing bed services but clearly such services are specifically the type of Post-Acute service which is the focus of your quest.

RESPONSE: Thank you for your comment. CAH swing beds do not currently trigger MSPB-PAC SNF episodes as they are not reimbursed under the SNF PPS and do not report the necessary information for payment standardization. Without being able to calculate standardized payment, it is difficult to make meaningful comparisons between CAH swing bed and other SNF facilities; for this reason, they are excluded from the MSPB-PAC SNF measure.



Name, Credentials, Organization, and Email of Commenter: Josh Luke, PhD, FACHE,

Adjunct Faculty, University of Southern California Sol Price School of Public Policy. Founder,

National Readmission Prevention Collaborative [email protected]

Type of Organization: Academic institution, professional association

TEXT OF COMMENT:

I have spoken to post-acute providers in 6 states, on both coasts and in the Mid West and developed the following thoughts on this great proposal.

1. I believe this is the most effective proposal yet in prompting post-acute providers to engage in value based care, and not focus entirely on maximizing utilization and referring to partner agencies (allowing patients to utilize multiple levels of care during the same discharge episode)

2. I believe that 45 days post-acute discharge measurement period instead of 30 should be given consideration for the following reasons.

- Patients can be discharged home after 14 days in a SNF and return 2 weeks later without a hospital stay as a means of usurping



- It would be more consistent with the stated goal of only one level of post-acute care being necessary, ie, a patient should not need home health if effective SNF care is provided

3. SNF's will begin not referring to Home health on discharge. I believe this is ideal as the goal is one level of PA Care should be enough. However, since one week in a SNF is financially equivalent to an entire 6 week home health episode, SNF's are already buzzing that they will discontinue discharging patients home with home health and in return the SNF can keep the patient one week longer.

4. For the reason stated above, for SNF, is it feasible to split the measurement in half, with 1) SNF LOS compared to the national SNF average and 2) Total PAC spending compared to average.

I hope this is helpful.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-1 and 22.

Regarding your comment on discharge practices, ultimately, facilities are responsible for providing appropriate care and follow-up for each patient. We expect that other outcome measures will identify patterns of inappropriate care.

Regarding your comment on splitting the measurement, the IMPACT Act mandates the development of a Medicare spending per beneficiary resource use measure. A separate measure would be needed to measure relative length of stay between providers.



Name, Credentials, Organization, and Email of Commenter: Debbie Baer, HIPAA Officer,

Risk Manager, Saint Mary’s Home of Erie [email protected]

Type of Organization: Retirement community

TEXT OF COMMENT:

After reading through the Draft Specifications for the MSPB-PAC Resource Use Measures draft, I would like to share a few comments. I agree that to determine and demonstrate quality, there must be measurement. The measurement definitions described in the draft are very complex. I feel that to rely on the results of these measurements and use them for comparison purposes, it is important to try it on a subset within each category of setting-specific providers. To make these the standards for which providers will be judged upon, the process needs to be tested to assure the elements of the calculations are in fact taking the appropriate data into consideration. A pilot program would be recommended to allow a review of the results for accuracy of the calculations, but more importantly – to evaluate if the results are accomplishing the goal of the project in providing meaningful information.



Again, thank you for the opportunity to provide comments.



Measure Set or Measure: LTCH

Name, Credentials, Organization, and Email of Commenter: Susan Klanecky, MSN, RN,

CCM, CRRN, Vice President of Patient Care, Madonna Rehabilitation Specialty Hospital

[email protected]

Type of Organization: LTCH

TEXT OF COMMENT:

Madonna Rehabilitation Specialty Hospital (MRSH) is located in Lincoln, Nebraska, and provides specialized programs of care to chronically and critically ill and medically complex patients who are Medicare beneficiaries. MRSH is pleased to present comments on the Medicare Spending per Beneficiary for the Post-Acute Care (PAC) Long-Term Care Hospital Quality Reporting Program (Required under the IMPACT Act).

While we support the measure’s goal of evaluating providers’ resource use relative to the resource use of the national median provider, we have identified several concerns regarding the ability of the measure to accurately reflect resource use differences across LTCHs. We discuss these concerns below.

Cross-Setting Comparisons

We appreciate that “each MSPB-PAC measure only compares providers within a given PAC setting” and that “different types of PAC providers are not compared to one another.” We echo the concerns of several of the panelists during the October 29 and 30, 2015 Technical Expert Panel on the MSPB-PAC measures that current data and methods do not permit comparisons across different types of PAC providers on spending over an episode of care. We agree that “given the lack of standardized assessment data, as well as inherent differences in payment systems and patients populations across PAC settings”, CMS should “undertake considerable research and gather substantial stakeholder input if the measures were to be adapted for this purpose in the future.”

LTCHs differ from other PAC settings in important ways. First, LTCHs, along with inpatient rehabilitation facilities, must meet the requirements of an acute care hospital. Second, LTCHs must have an average length of stay of more than 25 days. As a result, LTCHs treat higher severity cases than other post-acute care settings. In 2006, approximately 37% of LTCH cases grouped to the highest APR-DRG severity score, while this percent ranged from 4% to 7% for other post-acute care (PAC) providers. (Koenig et al. The Effects of Long-term Care Hospitals



on Outcomes, Utilization and Payments for Medicare Beneficiaries. November 7, 2013. Final Report prepared for the National Association of Long Term Hospitals.) Patients treated in LTCHs often possess multiple comorbidities and require specialized care. For example, 28.0% of LTCH patients with digestive system problems had at least three major complications or comorbidities compared to 2.2% of patients with digestive system problems in other PAC settings. (Lane Koenig, Berna Demiralp, Josh Saavoss, and Qian Zhang, “The Role of Long-term Acute Care Hospitals in Treating the Critically Ill and Medically Complex: An Analysis of Nonventilator Patients,” Medical Care 53 (7) (July 2015): 585) As another example, LTCHs frequently treat patients on prolonged mechanical ventilator with the purpose of weaning the patient. Few other PAC providers see such patients. We do not believe that claims data alone are sufficient to appropriately adjust or control for severity differences between LTCHs and other providers. Thus, we strongly support separate measures for each of the PAC settings.

Issues regarding LTCH interrupted stays.

The specifications do not discuss how LTCH interrupted stays will be treated. Therefore, we suggest that the measure development team clarify this issue.

Concerns with including short LTCH stays

We are concerned that the MSPB-PAC measure will hold LTCHs accountable for very short LTCH stays. For example, consider a patient that is discharged from a STCH to an LTCH, stays in the LTCH for a couple days, and is then readmitted to the STCH. The LTCH stay will trigger a new episode with the spending during the LTHC stay included as treatment services and the spending during the readmission to the STCH included as associated services. However, since LTCH are defined, in part, based on having an average length of stay of more than 25 days, these short stays may represent episodes where the STCH discharged the patient too early. The LTCH should not be accountable for the patient in these instances. We ask the measure development team to consider establishing a minimum number of days in an LTCH or other setting before an episode counts toward the MSPB-PAC measure.

Issues related to Risk Adjustment

In comparing between LTCH facilities, we are concerned that the current risk adjustment variables will not adequately capture patient differences. Without sufficient risk adjustment, differences in spending may be due to differences in patients’ clinical characteristics and may not be attributable to differences in resource use across providers.

a. We recommend adjusting for additional factors that play a significant role in determining resource use in the LTCH setting, such as prolonged mechanical ventilation, multiple organ failure (beyond those included in the variable interaction terms), and the number of days in an intensive care unit. We also recommend adjustment based on socioeconomic factors.

b. Even though a separate risk adjustment model will be estimated for episodes with the most recent institutional claim being a prior PAC claim, we recommend adding the type of prior PAC setting as a risk adjustor. Patients with a prior HHA claim and patients with a prior LTCH claim may differ greatly in terms of their acuity.


c. While we support estimating separate risk adjustment models or the six clinical case-mix categories described on pg. 21 and a separate model for LTCH site-neutral cases, we are concerned that the sample sizes in each of the clinical case-mix categories will lead to imprecise risk adjustment. We suggest that the measure development team present statistical analyses showing the effect of using these clinical case-mix categories on the precision of the predicted episode payments from the risk adjustment model as well as estimates of the reliability of the estimates.

d. While Medicare claims data are more readily available than other data sources, they may not capture finer distinctions across patients that may affect the patients’ outcomes and facility to which they are discharged. Therefore, a process to include assessment data in the MSP-BPAC measure calculations, once available, needs to be established and followed.

Concerns regarding the public comment process

While we appreciate that the comment period was extended by two days, the period provided for public comment is insufficient to provide complete comments regarding these complex measures that may have profound implications for MRSH. We encourage CMs and the measure development team to provide at least 30 days for the comment periods in the future.

In addition, the specifications do not provide enough detail for a convincing argument on some of the choices made by the measure development team. For example, collapsing readmissions for the same patient and provider within 7 or fewer days as part of the same treatment period may be reasonable, but the specifications provider sparse support for this decision in their appendix D. In addition to listing the types of empirical analysis conducted, we would find it helpful for the measure development team to report the results from these analyses so the public can make their own conclusions regarding the analyses.

We appreciate your consideration of these important issues.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-4, 9, 11, 19, 21, 34, 39, 41, and 42.



Name, Credentials, Organization, and Email of Commenter: Akinluwa (Akin) Demehin,

Senior Associate Director, Policy, American Hospital Association [email protected]


TEXT OF COMMENT:

On behalf of our nearly 5,000 member hospitals, health systems and other health care organizations – including over 3,300 institutionally based or affiliated providers of acute long-



term care, inpatient rehabilitation, hospitals with skilled nursing and extended care beds, hospital-based or affiliated home health agencies, and hospitals offering a spectrum of non-institutional services -- the American Hospital Association (AHA) appreciates the opportunity to comment on the post-acute care (PAC) Medicare Spending per Beneficiary (MSPB) measures. The PAC MSPB measures are being developed for future implementation in the Centers for Medicare & Medicaid Services’ (CMS) quality reporting programs for PAC providers, as required by the Improving Medicare Post-Acute Care Transformation (IMPACT) Act. The AHA and its members are deeply committed to enhancing healthcare value—that is, delivering better outcomes at lower costs. There is a clear need for standardized measures of cost, efficiency and resource use so that the field can accurately benchmark and track performance over time. However, we believe there are significant issues with the PAC MSPB measures that must be addressed before any of them are implemented in CMS quality reporting programs. In particular, we urge that the remainder of the measure development process be attentive to the following issues: Measure Testing. The AHA strongly urges that the measures be tested for reliability and validity, and that full information about measure testing be made publicly available prior to implementation. Furthermore, we urge that the measure undergo field testing with PAC providers – such as through a CMS-convened “dry run” – prior to implementation in any CMS programs. The draft measure document provide a variety of information about the measure cohort, episode length, exclusions and risk adjustment variables that are proposed for the MSPB measures. However, the draft specifications provide virtually no data that would enable the field to evaluate Acumen’s measure design decisions. For example, there are no descriptive statistics showing the distribution of performance by characteristics like bed size or urban/rural status. We also lack information on the level of statistical significance of the variables chosen for the risk adjustment model.

Given that the PAC MSPB measures will be publicly reported, it is imperative that they provide an accurate portrayal of provider performance. For this reason, Acumen and CMS must ensure that the measure is fully tested, and that the results of that testing are fully transparent so that all stakeholders have an opportunity to suggest meaningful improvements to the measure. Indeed, these data also would be expected to be submitted as part of the NQF endorsement process, and the AHA strongly recommends that all measures in CMS programs receive NQF endorsement. In addition, we recommend that Acumen and CMS conduct a “dry run” in which all PAC providers are given confidential preview reports of their performance prior to publicly reporting the measure. CMS has used dry runs in the past – including in its PAC quality reporting programs – for new measures in its quality reporting programs so that providers can become familiar with the methodology, understand the measure results, know how well they are performing, and have an opportunity to give CMS feedback on potential technical issues with the measures. Given that the PAC MSPB measures will mark the first time CMS is publicly reporting information about PAC resource use, we believe a dry run would be a crucially important step to enhancing the understanding and credibility of the measures.


Risk Adjustment. The AHA strongly urges Acumen and CMS to carefully evaluate the PAC MSPB measures’ risk adjustment approach. In particular, we are concerned that the measures do not adjust for patient functional status. We believe patient functional status is an important determinant of patient outcomes. Given that PAC providers are required by CMS to collect information on functional status as part of patient assessments, Acumen should explore whether it is feasible and not overly burdensome to providers to incorporate information from these assessments into the risk model. Sociodemographic factors. The AHA strongly urges Acumen and CMS to assess the PAC MSPB measures for the impact of sociodemographic factors on performance, and to incorporate adjustment as needed. As underscored by the National Academy of Medicine’s recent comprehensive review of the literature, there is a significant body of evidence showing the link between provider performance on outcomes, such as readmissions and cost, and socioeconomic factors like poverty, education and insurance status. We are concerned that without such adjustment, post-acute care providers caring for poorer and sicker patients will appear to perform worse on such measures than others treating a different patient population. Indeed, measures that fail to adjust for sociodemographic factors when there is a conceptual and empirical relationship between those factors and the measure outcome lack credibility, unfairly portray the performance of providers caring for more complex and challenging patient populations, and may serve to exacerbate health care disparities.

Thank you for the opportunity to comment.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-8, 11, 12, 30, and 31.

We appreciate your recommendation that the MSPB-PAC measures be submitted for NQF endorsement. The measures were presented to the NQF’s MAP Post-Acute Care/Long-Term Care Workgroup in December 2015 where they received a vote of “encourage continued development”. This vote was affirmed at a MAP meeting in January 2016. The finalized MSPB-PAC measures will be submitted for NQF endorsement.

COMMENT 15 of 45 Date: 1/29/2016

Measure Set or Measure: IRF

Name, Credentials, Organization, and Email of Commenter: Christopher A. Lee, MSPT,

FACHE, Vice President Rehabilitation, Madonna Rehabilitation Hospital – Lincoln

[email protected]

Type of Organization: IRF

TEXT OF COMMENT:



This letter is respectfully submitted on behalf of Madonna Rehabilitation Hospital – Lincoln (MRH-L). MRH-L is a 96 bed Inpatient Rehabilitation Hospital (IRH) located in Lincoln, Nebraska. MRH-L is one part of a Nebraska-based, not-for-profit, Catholic healthcare organization, referred to here as “Madonna”. Madonna is sponsored by Diocesan Health Ministries, a division of the Catholic Dioceses of Lincoln. Originally founded in 1958 by Benedictine Sisters whose mission was to “take care of the sick as Christ”, Madonna, the umbrella organization, has since grown to include two operationally separate hospitals (MRH-L and a LTCH), a skilled nursing facility, and outpatient services in Lincoln. The organization is also in the process of expanding into the Omaha metro market. All Madonna related healthcare entities are dedicated to the provision of rehabilitation care. Madonna is considered a local, regional, and national provider of comprehensive post-acute care services for MRH-L (our IRH) accounting for the majority of the regional and national draw. In response to regional and national needs, MRH-L has specialized programs to care for and help adults and children recover from brain injury, spinal cord injury, neurological diseases, and stroke.

Thank you for the opportunity to submit these comments on the draft measure specifications for the MSPB-PAC measure, particularly for the measure applicable to inpatient rehabilitation facilities (IRFs). The signatory of the letter, Christopher Lee, is a member of the Board of Directors of the American Medical Rehabilitation Providers Associations (AMRPA) and served on the committee that drafted the AMRPA letter on behalf of the industry. Therefore, MRH-L’s letter will closely follow the points made in the AMRPA letter.

MRH-L recognizes that the MRSP measure is required under the Improving Medicare Post-Acute Care Transformation (IMPACT) Act. As a result, our comments are designed to suggest improvements to the measure to ensure it achieves its intended goal of measuring resource use in a way that does not impose an undue burden on PAC providers and maintains access to these services for Medicare beneficiaries.

Section 2.5 Brief Description of Measures

MRH-L was pleased to see that a separate MSPB measure will be developed for each PAC setting and that comparison will occur between providers of the same type as opposed to comparisons between different types of PAC providers. We strongly agree that “setting specific measures allow for more meaningful comparisons to be made between providers than one single measure were calculated across all providers in all PAC settings,” and encourage the Center of Medicare and Medicaid Services (CMS) to continue its development of cross-setting PAC measures with this approach. We encourage Acumen and CMS to finalize these elements of the measure specifications.

In addition, we support the decision to use the national median based on the prospective payment system (PPS) base payment which does not include add-on payments or facility adjustors as the comparison benchmark.

Section 3.1 Episode Construction


The draft measure specifications escribe an episode window comprised of two parts – the treatment period and the associated services period. The treatment period includes the time period from admission to discharge at the PAC provider and the associated services period includes the services provided during the PAC stay and within 30 days of discharge from the PAC provider, with certain exclusions. Overall we support the inclusion of services provided during the PAC stay and for 30 days post-discharge in the episode window. However, we question if the proposed two-part composition of the episode (treatment period and associated services period) and their respectively attributed services creates a MSPB measure (s) that is unnecessarily complicated. Although the report states this bifurcation is necessary “because clinical exclusions of services in the treatment period may differ from clinical exclusion of services in the associated services period,” it does not provide any demonstrative examples. In the absence of further information, MRH-L would argue that those services excluded from the counting toward the MSPB episode in the treatment window, such as chemotherapy and other routine maintenance services, and planned admissions, should likewise always be excluded from counting towards MSPB episode if they occur during the associated services window.

The draft specifications state “The definition of PAC episodes allows episodes to overlap with hospital and other PAC episodes” (page 11) and “the full payment of all claims that begin within the episode window is counted toward the episode” (page 38). If we understand these statements correctly, one beneficiary’s trajectory of care could initiate multiple PAC MSPB episodes across different settings. An example of this would be a patient who is discharge from an IRF and at some point during the associated services period is admitted to another PAC provider such as a skilled nursing facility (SNF). In that scenario, the spending from the SNF stay will trigger a new MSPB episode for the SNF (due to SNF admission) and will also be attributed to the discharging IRF’s associated services period. In other words, the Medicare expenditures for a stay at one PAC setting will be associated with two MSPB-PAC episodes .This form of attribution is not inherently problematic and MRH-L supports this construction of the MSPB-PAC measures, as long as the measure remains a comparison of providers of the same type (e.g. IRFs to IRFs) as currently proposed. If however, at some point in the future CMS intends to compare different type of providers (e.g. SNFs to IRFs) this type of attribution would be inappropriate because it “doubly” describes one Medicare expenditure (the SNF stay in the example above) to two PAC providers. We request that Acumen and CMS clarify that this type of attribution will only occur while comparisons are taking place between PAC providers of the same type as opposed to across PAC providers.

Section 3.1.1 Opening (Triggering) Episodes

MRH-L is concerned with the treatment of readmissions within a seven day window of discharge as outlined in this section. Specifically, we are concerned that the readmission counts towards the original MSPB episode only if the patient is readmitted to the same IRF. Instead, we believe that the readmission should be attributed to the original IRF regardless of whether the patient returns to the original IRF or another IRF. In fact, we believe this form of attribution should occur across PAC providers so that the PAC MSPB measure does not unintentionally incentivize provider behaviors that are not in beneficiaries’ best interests.

We are also concerned about setting a threshold, in this case seven days, as we believe it may also lead to unintended consequences and efforts to manage the length of stay in a way that


avoids attribution for the discharging provider. We encourage Acumen and CMS to reevaluate whether such as [sic] threshold is appropriate or necessary.

Section 3.1.5 Step 5: Excluding Clinically Unrelated Services

The draft measure specifications note that certain services are excluded from the episode window (either the treatment period or associated services period) because they are clinically unrelated to PAC care and/or because PAC providers may have limited influence over certain Medicare services delivered by other providers. We agree that certain exclusions are appropriate and necessary particularly to avoid unintended consequences as outlined in the draft document. Therefore, we recommend that costly and routine care not related to the IRF stay, such as dialysis or chemotherapy, be excluded from the IRF MSPB episode.

Several examples of excluded services are provided throughout the draft measure specifications including on Table 3 (page 16) and in Appendix A. However, the document states while lists of clinically assessed services have been developed for each PAC setting by clinicians from CMS and the measure development contractor these lists are being further refined by these clinicians as well as an independent group of clinicians. Without a degree of certainty as to what exclusions are under consideration, it is challenging to provide constructive feedback to CMS and its contractors on whether such exclusions are appropriate. MRH-L supports the exclusions specified in Appendix A, Table A-4 and recommends CMS adopt them.

Section 3.2.1 Implementing Episode-Level Exclusions

MRH-L supports the four beneficiary exclusions listed in the draft specifications including beneficiaries enrolled in Medicare Advantage, beneficiaries not continuously enrolled in fee-for-service (FFS) for the lookback period, beneficiaries whose primary payer is not Medicare, and beneficiaries who died during the episode. In addition, we think that patients who are discharged against medical advice (AMA) should also be excluded because a PAC provider is unable to control/manage those patients’ downstream trajectory of care and associated costs.

Section 3.2.2. Risk Adjustment Approach

One of the risk-adjustment factors CMS and Acumen intends to use are 70 Hierarchical Condition Categories (HCCs). However, MRH-L believes risk adjustment for patient severity in IRFs is more appropriately accomplished by using case mix groups (CMGs) rather than HCCs because HCCs were originally developed for risk-adjustment in the Medicare Advantage program and are based on ICd-9-CM codes. In summary, CMS and Acumen should use the CMGs rather than HCCs for IRFs.

In addition, we believe the measure should be risk adjusted for socioeconomic and sociodemographic status (SES/SDS). Short of incorporating a fleshed-out SES/SDS risk-adjustor, this could be accomplished through the use of claims data by using a beneficiary’s dual eligible status as a proxy for socioeconomic/demographic status.

Section 3.2.3 MSPB Measure Calculation

We appreciate that the measure calculation is relatively complex in part to ensure the fairness of the calculation. But we believe it is important that IRFs be able to replicate the calculation and


verify the MSPB value assigned to them. As currently constructed the calculation is too complex to enable them to do so. We also believe that a single value indicating the average spending provides little actionable data upon which PAC providers such as IRFs can utilize to manage spending more effectively. MRH-L recommends CMS be transparent with the data it uses to calculate this measure and that CMS makes the data available, with the necessary elements (such as exclusions, etc.), to providers so that providers have a working knowledge of the measure, its calculation, and the provider’s standing.

The draft measure specifications also do not delineate if or how a PAC provider who disagrees with the value assigned to them through the measure calculation may appeal the determination. It is critically important that such a mechanism is available to providers.

Finally, through a post-payment audit, the provider may be required to return funds to CMS that were part of the MSPB calculation. If such an event occurs, the provider’s MSBP value should be recalculated to reflect the lower spending.

Conclusion

In closing, MRH-L appreciates the opportunity to provider our recommendations for improving the structure of the MSPB measure. Your consideration of these important issues is also greatly appreciated.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-4, 5, 11, 15, 16, 18, 26, 28, 30, 37, and 48.

Regarding your comment on the episode window, we mostly expect that services such as chemotherapy and planned admissions, if excluded in the treatment period, will also be excluded throughout the associated services period. We do however retain the separation of the two periods to ensure that the MSPB-PAC measures are constructed in such a way to accommodate the possibility that service-level exclusions may be appropriate (or inappropriate) in only the treatment period. For example, it may be appropriate to exclude an infection occurring 3 weeks after discharge but include it if it occurs during the treatment period. There may also be some instances where a patient is receiving physical therapy outside of the PAC setting. As those should be provided by the PAC, these services should not be excluded during the treatment period, but they may potentially be excluded in the post-treatment period. It may also be important to account for this difference between treatment and associated services periods in the event of a future policy change that affects provider practices.



Name, Credentials, Organization, and Email of Commenter: Cheryl Phillips, MD, Senior VP

Public Policy and Advocacy, LeadingAge [email protected]




TEXT OF COMMENT:

The members of LeadingAge and affiliates touch the lives of 4 million individuals, families, employees and volunteers every day. The LeadingAge community (www.LeadingAge.org) includes 6,000 not-for-profit organizations in the United States, 39 state partners, hundreds of businesses, research partners, consumer organizations, foundations and a broad global network of aging services organizations that reach over 30 countries. The work of LeadingAge is focused on advocacy, education, and applied research. We promote home health, hospice, community- based services, adult day service, PACE, senior housing, assisted living residences, continuing care communities, nursing homes as well as technology solutions and person-centered practices that support the overall health and wellbeing of seniors, children, and those with special needs.

LeadingAge continues to support the principles and objectives of the Improving Medicare Post-Acute Care Transformation (IMPACT) Act of 2014, and remains committed to working with CMS and post acute care industry partners to see the measures required in the law are implemented as intended. However, we wish to express our concern that the timeline for meaningful review of complex measures is not nearly long enough. A period of longer than two weeks should be provided for careful consideration and comment, given the goal of developing uniform measures that can eventually apply across PAC settings. This measure requires continued development and further analyses before implementation.

Specifically related to the MSPB-PAC, there are five points that we raise for your consideration.

Resource use versus efficiency

LeadingAge believes for consistency purposes the MSPB-PAC should be referred to as a resource use measure and not implied as an efficiency measure. Throughout the draft specifications, the terms resource use and efficiency are used almost interchangeably when a real distinction exists between them. As defined the MSPB-PAC compares average provider costs to national median costs. This is in absence of any quality outcomes tied to resource use, which are a key component of efficiency measures that seek to ascertain value for cost. The MSPB-PAC should not be implied to be an efficiency measure. We have concerns that providers could reduce the delivery of necessary services to appear more efficient, certainly a negative unintended consequence.

Components of episodic costs

LeadingAge believes that an accurate determination of Medicare spending should take into account Medicare Parts A, B, and D expenditures. The omission of pharmaceutical expenditures should be reconsidered. With the development of a resource use measure and the goal of reducing variation in PAC spending including pharmacy costs makes sense.

Home health episodic treatment

LeadingAge is concerned that the episode length used for standard home health and HHA-LUPA episodes should not inherently be sixty days. For cases when the home health episode ends before day sixty, the treatment period should end at discharge. The LUPAs are four or fewer billable visits; roughly nine percent of episodes and one percent of payments are from LUPAs. While last year an increase was given for LUPA payments, the LUPA is not case mix adjusted

http://www.leadingage.org


like a standard home health episode. The treatment period for all categories of post-acute care episodes should end at discharge.

Additionally, there is a difference in utilization and scope of services provided for post-acute home health compared to the community admitted users of home health. Only about 34 percent of total home health visits are post acute. Non-post acute episodes tend to have a greater share of lower cost home health aide visits. In the risk adjustment described in the draft specifications community users of home health are assigned the lowest priority. We suggest that only post-acute home health episodes are included in the measure to be uniform across settings as all skilled nursing facility Medicare days are post acute. This aligns with the IMPACT goals of unifying the treatment of post-acute care irrespective of setting.

Consistency in statistical representation of average

LeadingAge notes a lack of consistency in the definition of the numerator and denominator of the MSPB-PAC. Average risk-adjusted episode spending is in the numerator while the episode-weighted national median spending is in the denominator. Are the average in the numerator intended to be the median episodic spending or is this a comparison of mean episodic spending over median spending? We are in favor of the use of median spending to account for variation present particularly with low-volume providers as it does a better job of limiting skewness.

Risk adjustment

LeadingAge is supportive of the efforts to risk-adjust for patient characteristics but believe that the included characteristics leave out important variables that relate to resource use. One particular category would be socio-economic characteristics, which have shown to be related to resource use for medical care services. For those beneficiaries who can pay for non-skilled services out of pocket, the result will be fewer expenses even for Medicare. As an example, the rehab patient who can afford an in-home care worker to assist with ADLs and IADLs and therapy exercises may be discharged sooner that a person who cannot pay for such services and thus needs to be at a higher level of function prior to discharge from their post-acute setting. With CMS’ recent focus on improving readmission rates for beneficiaries of racial and ethnic minority groups, the inclusion of social determinants of health should be included throughout the post-acute measures as well.

Again, LeadingAge appreciates the opportunity to comment on this draft measure. We hope our comments will be helpful to you.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-3, 8, 9, 12, 20, 23, 29, 30, and 47.




Name, Credentials, Organization, and Email of Commenter: Cynthia K. Morton, MPA,

Executive Vice President, National Association for the Support of Long Term Care

[email protected]


TEXT OF COMMENT:

The National Association for the Support of Long Term Care (NASL) appreciates the opportunity to comment on the draft specifications for the Medicare Spending Per Beneficiary—Post-Acute Care (MSPB-PAC) resource use measures as mandated by the IMPACT Act. NASL is a national trade association representing providers of ancillary services to long term and post-acute care (LTPAC) settings. NASL’s members include rehabilitation therapy providers that employ physical therapists, occupational therapists and speech-language pathologists who provide therapy services to patients in skilled nursing facilities (SNFs), and other long term and post-acute care (LTPAC) settings. In addition, NASL represents suppliers of durable medical equipment, suppliers of enteral nutrition and developers of health information technology (IT) with full clinical and point-of-care IT systems. NASL members have reviewed the draft as extensively as possible in the time allowed and we have prepared the following comments. We also support the comments submitted by the American Health Care Association. Before we discuss several specific issues with the draft measure, we must state two serious concerns we have regarding the presentation of the draft measure to the public. First, the time frame for review and comment of the draft measure-- approximately two weeks and two days--was overly condensed, in our view. Second, we believe there is not enough information in the draft report to determine the validity of the measure. While NASL highlights the following concerns and comments, we find it difficult to comment effectively when there is such a high degree of uncertainty as to the workings of the measure as well as the intent behind it. Areas of Concern Exclusion of services clinically unrelated to the PAC treatment

• We recommend that far more clarification should be provided regarding routine and other services that will be excluded during the treatment period and during the associated services period. The draft measure would exclude routine costs that are unrelated to the triggering event that caused admission to the SNF. Does that extend to services included in SNF Part A RUGs? For example, if a patient requires oxygen for the treatment of COPD and the triggering event is not related to COPD, would all costs related to the treatment of COPD be excluded because they are unrelated to the triggering event – despite the fact that the SNF would be required to provide the oxygen and payment for this treatment would be included in the RUG payment? The SNF must incur these costs when the services are medically necessary, and if they are excluded from the episode, this will lower and distort the costs attributed to the episode.

• In addition, the exclusion provisions in the draft are not clear as to whether Part B therapy related to the routine medical care would be excluded. For example, a patient



requires physical therapy related to his diabetes, but the triggering event that required admission to the SNF was unrelated to diabetes, would this therapy be excluded?

These concerns point to a very important issue. Why would costs routine or otherwise be excluded from the cost of the stay? These costs were borne by the provider, were required to be provided by statute and should be part of any cost per episode calculation. To exclude costs borne by the SNF would artificially lower the cost totals and impact the measure. SNF Patients Often Have Multiple Diagnosis

• Often patients admitted to the SNF have multiple medical complexities and or comorbidities, which do not appear to be addressed in the draft. It is unclear if the measure connects a diagnosis with the triggering event—the admission to the SNF. The measure does not appear to take into account that there could be more than one medical problem or trigger that requires admission to the SNF. In fact, the measure does not seem to indicate how the trigger will be identified or identified by using a Plan of Care or other such tool that shows the full range of reasons why the patient was admitted to the SNF. There is some mention of diagnosis with regard to exclusions in Table 4 on page 18. While there is an admission primary diagnosis on the claim, that diagnosis can sometimes be resolved quickly and subsequent diagnoses on the claim can support the skilled stay and episode. Will all of these diagnoses be taken into consideration for the process of inclusion or exclusion of associated services?

Statutory Requirements of the IMPACT Act of 2014 The draft measure does not appear to meet the clear intent behind the resource use provisions contained in the IMPACT Act:

• The overall design does not provide resource use information that allows comparison of different Medicare post-acute care providers, which was the primary intent behind the resource use provisions of the statute.

o Rather, the measure is designed to compare resource use within each segment of the PAC spectrum, in isolation of the other PAC providers. It only allows for comparison of SNF to SNF, LTCH to LTCH, HH to HH, IRF to IRF. Most of this is already feasible under established CMS quality reporting and payment policies.

Measure Structure Designed Without Testing

• The draft report does not indicate whether the measure has been tested, and if it has not there appears to be no plan for testing. Also, the draft report mentions contractor-based clinicians who still are developing lists of exclusions, which indicates that work is not complete on this measure. As we stated in our general comments, much more detail needs to be provided and clarified regarding the exclusions.

Established CMS Billing Policies Are Disregarded

• The proposed measure does not appear to take into account the established Medicare timely filing process that allows Medicare providers up to 12 months to submit claims for payment of services. In the draft, costs for services and products outside current Medicare payment policy (Physician/Supplier and DME) are attributed to a provider and included in the calculation of the measure. If the providers that supply services such as Part B and


DME services have one year to file their claims, at what point will the costs from these claims be available in order to incorporate into the measure?

• Provider liability can be unknown for the duration of the episode window and could extend for months post discharge from the entire episode window while waiting for all claims (both from the attributed provider and claims from the associated service period) to be adjudicated. In addition, claims determinations can be appealed, which could be a multi-year process. At what point will all final claims be available for use in calculating the measure?

Responsibility for Calculating the Measure

• The draft measure gives no indication where or when the measure calculation will occur. It is unclear how many episodes will go into the formula, i.e. a year’s worth, a month’s worth? Because the entity calculating the measure needs access to all claims, we assume that a Medicare Administrative Contractor will calculate the measure? Also, will a DME MAC provider DME claims to a MAC in order to include DME claims in the calculation?

Attributing Costs Accurately • Costs borne by the attributing provider and costs during the associated service period that

could be borne by another provider are included in the measure formula. Thus, one provider’s costs will be added to another provider’s episode window. The draft is silent as whether the attributing provider will have an opportunity to examine costs attributed to the episode window to check for accuracy. Depending on the time horizon of the measure, presumably hundreds of episodes or thousands of episode windows will be included in a measurement calculation. How will a provider be able to ensure that costs were attributed accurately? What kind of transparency will there be with the calculation? Will the provider receive a report showing the data that goes into the calculation before it is made public? What if information needs to be changed because it is inaccurate? How can a provider ensure claim information from associated services is accurate when the provider does not have access to those claims in some situations, such as with DME?

Provider Collaboration

• We do not see how the measure creates incentives for provider collaboration. The measure is a very complex calculation and there is no information in the draft as to where it will be handled or how transparent the process will be. If providers do not have access to the calculation, then they cannot attempt to influence it by collaboration with providers of associated services.

• We also question whether this measure incentivizes providers to stint on care and shift costs onto another PAC provider in order to improve a provider's efficiency score. For example, orthotics and devices such as a bone stimulator would be paid under the hospital stay if ordered when the patient was in an acute stay, even if the device is ultimately delivered when the patient is in the SNF. If the acute care provider delays obtaining orders for those items until the patient has been admitted to an SNF, the cost of those items would then fall on the SNF. This would shift the cost from the acute hospital to the SNF, thus driving up the cost of care provided in the SNF and decreasing their efficiency rating even though the SNF did not have full control over these expenses. It is


difficult to tell from the exclusion rules whether this type of item would be excluded and whether it should be excluded? Another example is a patient with a swallowing issue that is not identified in the acute hospital stay because the patient is not evaluated by a Speech Language Pathologist. When the patient is admitted to the SNF and the swallowing issue is identified the patient may require additional testing such as a Modified Barium Swallow or FEES. The MSPB measure actually could promote stinting of necessary care in one setting and shifting the costs of that care from one provider to another.

NASL is proud to represent the ancillary services that play an important, supportive role in the care of Medicare beneficiaries who require services in LTPAC settings and appreciates the opportunity to comment on the draft measure.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-5, 6, 7, 8, 9, 11, 14, 15, 16, 21, 23, 26, and 48.

Regarding your comment on SNF patients with multiple diagnoses, the episode trigger is independent of the diagnosis but rather the start of care within each PAC setting. A patient's clinical characteristics are accounted for through risk adjustment.

Regarding your comment on responsibility for calculation, the MSPB-PAC measures will be calculated by CMS using Medicare administrative claims data for a period to be specified as a part of the proposed rule for the LTCH, SNF, IRF, and HHA QRPs.



Name, Credentials, Organization, and Email of Commenter: Sharon L. Dunn, PT, PhD,

OCS, President, American Physical Therapy Association [email protected]


TEXT OF COMMENT:

On behalf of our 93,000 member physical therapists, physical therapist assistants, and students of physical therapy, the American Physical Therapy Association (APTA) respectfully submits comments regarding the draft specifications for the Medicare Spending Per Beneficiary Post-Acute Care (MSPB-PAC) Resource Use Measure as prepared for the Center for Clinical Standards and Quality within the Center for Medicare Centers for Medicare & Medicaid Services (CMS). Physical therapy is an integral service provided to Medicare beneficiaries in all post-acute care settings. Physical therapists furnish medically necessary services to patients to improve their overall health, function and to optimize their quality of life. Across the post-acute care settings, physical therapists provide physical therapy services to patients through a plan of care to engage and optimize the patient’s participation in achieving shared goals of improved functional performance, reduced risk of injurious falls, and reduced



risk of acute hospitalization thereby promoting long-term health and wellness. Physical therapists provide an examination that includes the history, systems review, and tests and measures to determine the patient’s therapeutic, rehabilitative, and functional status and any environmental factors that may impact the patient’s activity and/or participation. Through the evaluative process, the physical therapist develops a comprehensive plan of care to achieve the goals and outcomes of improved function. As mandated by the IMPACT Act, CMS must develop measures that address the total estimated Medicare spending per beneficiary for post-acute care providers. It is also stipulated that these measures should align with the hospital MSPB measure. APTA supports and understands the rationale for developing measures that better capture the cost and resource use of care in order to improve care. Physical therapists are committed to providing high-quality, timely care and to the promotion of evidence-based and patient-centered practice. Therefore, we ask that you carefully consider APTA’s comments and recommendations as articulated below. RECOMMENDATIONS First, we would like to express our appreciation to CMS and Acumen for its commitment to this important project, but we strongly recommend that this work is incorporated into the larger mandate of IMPACT which is the creation of standardized outcome measures across the respective post-acute care settings. This step is necessary to ensure that the totality of post-acute care is taken into account. It will also help to ensure that all post-acute care settings are assessed equally and patients are not unfairly steered to receive care in one setting over another based on flawed information regarding resource use and costs. One MSPB measure, in isolation with the limitations of claims based data, cannot define post-acute care. Therefore, it is imperative that this measure is used in concert with other measures to more fully define the scope of post-acute care services. The final MSPB measure should be used to effectively analyze the necessity of post-acute care services and not to merely make payment cuts.

Second, we support and advocate for the adoption of a MSPB measure that first compares cost and resource use within each provider type and not across the different post-acute care provider types. As evidenced by the PAC-PRD Demonstration Report and Med PAC studies, there are significant resource variations across the post-acute care settings that need to be adequately addressed before cross-setting comparisons can be achieved. By using the MSPB measure within each provider type, this will give each post-acute care setting the opportunity to test and assess the measure for appropriate refinements to address issues within the care setting. After this step has been taken, CMS will then have more information in which to appropriately develop and implement a cross-setting MSPB measure in tandem with the development and implementation of a unified post-acute care payment system. Third, we recommend that Acumen address resource use and cost for durable medical equipment (DME) in the development of this measure. APTA is concerned about how DME costs could be skewed if a provider has a disproportionate share of a certain patient population. This is particularly true for settings that treat a high volume of patients with obesity, amputations, and CVAs (hemiplegia/paresis) for example. These providers often have higher cost for DME that are outside of their control. If these items are not accurately captured in the risk adjustment of the MSPB measure, APTA is concerned that these providers could be unfairly penalized. We are


also concerned that providers could be reluctant to treat patients who may be predicted to have higher costs in areas such as DME. Fourth, APTA questions the “90 days prior to the episode trigger”. We are concerned that the 90 day timeframe is not sufficient to capture complex patients with comorbidities and chronic conditions. Therefore, we advocate for a 180 day window. We also ask that Acumen clarify whether the 90 day window is from the triggering post-acute care admission or from the triggering hospital admission.

Last, APTA is concerned about the proposed construct of the national median as the benchmark for the MSPB measure. There is great variance is cost across geographic regions due different practice patterns, community resources and patient populations. In addition, the effect of Medicare Advantage (MA) plans play a significant role in cost across the geographic regions which may present a very different Medicare Fee-For-Service population than in other regions where MA plans are not as prevalent. Therefore, we recommend that CMS employ the use of state or regional benchmarks first in order to better capture costs before moving to a national benchmark median. In conclusion, APTA appreciates the opportunity to review these draft specifications as well as the stakeholder input that was provided during the recent technical expert panel, but we believe that this measure should be thoroughly tested and vetted within the post-acute care community prior to implementation. Therefore, we respectfully request that CMS undertake a voluntary testing period for at least six-months with the intent to make final refinements based on the issues that arise during this testing phase. APTA is more than willing to be a resource to recruit providers and facilitate the distribution of information during this testing period. Once again, we thank CMS for the opportunity to comment on these draft measure specifications. RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 4, 6, 8, 12, 43, and 46.



Name, Credentials, Organization, and Email of Commenter: Cheryl A. Burzynski, MSN,

RN, NE-BC, President, McLaren Bay Special Care [email protected]


TEXT OF COMMENT:

McLaren Bay Special Care Hospital (MBSC) is grateful to have the opportunity to present comments on the DRAFT Specifications for the Medicare Spending per Beneficiary Post-Acute Care (MSPB-PAC). MBSC is an Acute, Long-Term Care Hospital (LTCH). We opened our



LTCH in 1994 in a small, micro-urban community within East Central Michigan with a high population of elderly. Many of the residents in Bay City have adult children that had to seek jobs outside of this area and out of state due to a decline in the automotive industry years ago.

We have sincere concerns regarding the ability of the measure to accurately reflect resource use differences across LTCH’s. As stated above, we are considered a small, urban hospital, yet our city now has a census less than 33,000, our county around 100,000, yet we do not have the distinction of rural.

We are happy that “each MSPB-PAC measure only compares provides within a given PAC” and that “different types of PAC providers are not compared to one another.” I feel that “given the lack of standardized assessment data, as well as inherent differences in payment systems and patient populations across the PAC setting,” CMS needs to “undertake considerable research and gather substantial stakeholder input if the measure were to be adapted for this purpose in the future.”

LTCH’s differ considerably from one geographic to another let alone from other PAC settings. LTCH’s are licensed as acute care hospitals, and need to follow the rules for hospital level of care. For our new LTCH qualifying cases, we need to have the patient come from a Subsection D Hospital, have been 3+ days in an ICU in a Subsection D Hospital, or been on a ventilator from a Subsection D Hospital in the LTCH Hospital for 96 hours prior to successful weaning. From our north and east regions, we are surrounded by rural hospitals that do not qualify as appropriate ICU admission as noted above. We do not believe that claims data alone are sufficient to appropriately adjust or control for severity differences between LTCHs and other providers. We strongly support separate measures for each of the PAC settings. In addition, CMS needs to make it very clear when the data becomes available for public display that it is not appropriate to compared MSPB-APC measure values across PAC Settings.

MSPB is NOT a Quality Measure: On page 4 of the measure specification, we do not agree with the statement, “Since the design of the MSPB-PAC measures, post-treatment costs may serve as an indicator of the quality of care provided during PAC care, in that higher quality PAC treatment may yield lower post-treatment costs.” Costs by themselves are not an indication of quality and should never be interpreted as such. If you do this, the result could cause negative consequences for Medicare beneficiaries. Costs must be interpreted in combination with clinical outcomes. The MSPB does not include clinical outcomes, so it cannot by itself reflect quality.

Throughout the DRAFT document, it is stated that the MSPB-PAC measures reflect provider efficiency, we disagree again, and ask that Acumen and CMS refer to the MSPB-PAC measures as resource use measures and no more.

A serious concern of ours is that issue of Interrupted Stays in the LTCH. The specifications do not discuss how LTCH interrupted stays will be treated. While services received for interrupted stays for three days or less are the responsibility of the LTCH, services received during the interrupted stays of more than three days are included in the claim for the short-term acute care hospital (STACH), IRF or SNF. If the LTCH is a site-neutral episode, the interrupted stay is only counted during the treatment period and the specifications do not clarify if the STACH,


IRF, or SNF claims will be used to capture treatment services during the interrupted stay. Please clarify for the LTCH’s how these interrupted stays will be treated.

RISK ADJUSTMENT CONCERNS:

Since LTCH’s vary significantly due to geographic locations and surrounding STACH hospitals, we are concerned that the risk adjustment variables will not adequately capture patient differences. Without sufficient risk adjustment, differences in spending may be due to the differences in patients’ clinical characteristics and may not be attributable to differences in resource uses across providers.

1). Recommendation: adjusting for additional factors that play a significant role in determining resource use in the LTCH setting, such as prolonged mechanical ventilation, multiple organ failure (beyond those included in the variable interaction terms), and the number of days in the ICU. Also recommended is adjustment based upon socioeconomic factors.

2). Even though a separate risk adjustment model will be estimated for episodes with the most recent institutional claim being a prior PAC claim, it is recommend to add the type of prior PAC setting as a risk adjustor. Patients with a prior HHA claim and patients with a prior LTCH claim may differ greatly in terms of their acuity.

3). While we can support estimating the separate risk adjustment models for the six clinical case –mix categories described on page 21, and a separate model for the LTCH site-neutral case, we are concerned that the sample sizes in each of the clinical case-mix categories will lead to imprecise risk adjustment. Suggestion: the measure development team present statistical analyses showing the effect of using these clinical case-mix categories on the precision of the predicted episode payments from the risk adjustment model as well as estimates of the reliability of the estimates.

4). While Medicare claims data are readily available than other data sources, they may not capture the finer distinctions across patients that may affect the patients’ outcomes and facility to which they are discharged. Therefore, a process to include assessment data in the MSPB-PAC measure calculations, once available, needs to be established and followed.

MBSC and other LTCH’s are concerned that the utilization of this specific measure will not allow us the capability to continually calculate this measure internally. We believe that the utilization of an LTCH specific MSPB measure should not be shared at this time with the public or published on the upcoming LTCH Compare Website.

CONCERNS REGARDING THE PUBLIC COMMENT PROCESS

While the extra two day extension for the comment period occurred, that is really insufficient time for public comment. This is complex information with a profound effect on LTCH’s and MBSC.

We would ask for at least a 30 day comment period in the future. Analysis takes time and multiple individuals’ input.


The specifications provided for comment do not provide enough detail for a convincing argument on some of the choices made by the measure development team. As an example, collapsing readmissions for the same patients and providers within 7 or few days as part of the same treatment period may be reasonable, but the specification provide sparse support for this decision in Acumen’s Appendix D. In addition to listing the types of empirical analysis conducted, it would be helpful for the measure development team to report their results from these analyses so that the public can make their own conclusions regarding the analyses.

Please feel free to contact me about these comments.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 3, 4, 5, 9, 11, 13, 15, 19, 30, 34, 39, 41, 42, and 45.



Name, Credentials, Organization, and Email of Commenter: James R. Prister, President &

CEO, RML Specialty Hospital [email protected]


TEXT OF COMMENT:

RML Specialty Hospital (RML) is pleased to have the opportunity to present comments on the Draft Specifications for the Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) Resource Use Measures, Provided for Public Comment, January, 2016. RML is a freestanding hospital (with 2 locations) licensed in the State of Illinois and recognized by Medicare as a long term acute care hospital. RML is a 501(c)3 not-for-profit limited partnership whose current members are Loyola University Medical Center and the Advocate Healthcare Network. RML’s clinical focus is on ventilator weaning (respiratory), complex medical, and wound services. Because of these programs, RML has historically maintained a very high case-mix level. During the last 12 months, our average case-mix fluctuated between 1.4 - 1.5 for Medicare patients. Our high case-mix level continues even after the significant case-weight decreases in the LTC-MS-DRG system from previous years. Patients are referred to RML from approximately 65 hospitals in Illinois. Most patients are normally transferred from ICUs, critical care units, burn units, and step-down units. The purpose of this letter is to provide some general input, express concerns, and seek clarification regarding several items contained in the above Draft Specifications. RML appreciates Acumen’s and CMS’ thoughtful consideration of our comments and suggestions. As a general statement, RML is very concerned about the utilization of this specific measure as we will be unable to continually calculate this measure internally. The only way to receive this measure will be to obtain it from CMS. This is not a good process for providers to use to



improve outcomes/results. Data must be readily available and current! Additionally, we believe there are several assumptions on the calculation of the metric which needs further clarification and study. Lastly, we believe the utilization of an LTCH specific MSPB measure should not be shared at this time with the public or published on the upcoming LTCH Compare website. We strongly suggest that this measure, as it is Medicare “spending” and not a patient out of pocket metric, could lead to significant and inappropriate decisions on where patients will be discharged. We are very appreciative of the fact that CMS has identified that these post-acute measures should not be compared across the various post-acute settings.

Before we make specific comments about the proposed MSPB-PAC, we believe it is imperative that we share some of the current perceptions and issues we are encountering from short stay hospitals and how they are utilizing their own MSPB statistics. As you are very aware, the short stay hospital MSPB statistic includes 30 day post-discharge costs which would include an LTCH stay if there was one. An LTCH stay is very likely going to extend beyond the 30 day discharge window from the short stay hospital. However, the LTCH’s entire stay reimbursement is included in the short stay MSPB metric. We are aware of a number of short stay hospitals in the Chicago area that are starting to make discharge decisions based on their MSPB associated with the various post-acute discharge destinations. Cases should go to an LTCH because of the clinical benefit associated with the LTCH, not the spend impact. RML strongly suggests that CMS reconsider the inclusion of LTCH appropriate cases in short stay hospital MSPB calculations. Alignment of care and coordination of care should not be based solely on an MSPB indicator. The MSPB metric may be good for average cases, however we are skeptical of the risk adjustment for patients that are high-cost outliers. Many of the cases that are admitted to RML from short stay hospitals are outliers at the short stay hospital. Medicare uses outlier pools in both the short stay hospital reimbursement system and the LTCH reimbursement system for cases that do not fit the average. It does not appear that either the short stay MSPB nor the proposed MSPB-PAC does enough to risk adjust for these outlier patients. It is imperative that CMS provide access to as real time data as possible in order for LTCHs to prepare for and potentially manage this metric in the future. Having easy access to post-discharge “Medicare spend”, will be vitally important. We suggest this information be made available to providers on a real time basis.

INTRODUCTION STATEMENT It is identified that the purpose of the MSPB-PAC measure is to support public reporting of resource use in all four post-acute care settings as well as to provide actionable, transparent information to support PAC providers’ efforts to promote care coordination and improve the efficiency of care provided to their patients. We urge CMS to utilize this measure with great caution in the post-acute setting. Many individuals in the public would assume that the spending per beneficiary metric may be the cost they will pay vs. the amount that Medicare would spend for the treatment provided in that particular setting. This metric should not drive a patient’s decision on which setting or provider should be utilized. We support the concept of providing transparent information if it can be well defined and understood by the public.


The second concern under this section is whether this metric provides actionable information. Throughout the explanation of the various MSPB-PAC definitions, it is identified that the contractor used internal resources to articulate inclusions, exclusions, etc. An individual provider does not have the level of research resources used by the contractor to develop the metric. If the metric definition changes every year, then it becomes unmanageable for a provider.

It is suggested in the Introduction that there is a belief that the design of the MSPB-PAC measure may serve as an indicator of the quality of care provided during PAC care and that high quality PAC treatment may yield lower post treatment costs. This is a wonderful statement, however we must stress that this measure does not take into account any patient outcomes and it likely does not reflect actual patient complexities. We strongly concur and support CMS’ statement that PAC providers should not be compared across PAC settings. Acumen identified there will be separate LTCH appropriate vs. LTCH site neutral metrics. We support this determination. We also suggest that Acumen consider some additional distinctions that could be helpful in the LTCH setting. Specifically, the differences between free standing and HIH facilities. In theory, the Medicare spend may be the same between these two settings. However, Acumen should test whether there should be an adjustment to the metric because of the inherent cost difference between an HIH and a free standing LTCH. Acumen identifies that they have attempted to risk adjust for dialysis as part of this measure. As we know, there is no separate reimbursement for a patient on dialysis in the LTCH setting. We are including information that we have provided to CMS in the past regarding our dialysis utilization at RML. Specifically, we are very concerned that a patient on a vent and receiving dialysis has the same “Medicare spend” as a patient who is just on a ventilator. The sensitivity of the LTC-MS-DRG system does not incorporate the dialysis treatment nor is there an additional payment by Medicare for patients on dialysis. We strongly request that this situation be re-evaluated and an additional risk adjustment be made for an organization like RML that has between 18% - 20% of our entire patient population on dialysis vs. another LTCH who has 0% - 2% of their entire patient population on dialysis. We do not believe the current risk adjustment is sufficient or appropriate at this time. HIGH VOLUME DIALYSIS LTCHs In each of the last several years, RML submitted comments to CMS requesting that consideration be given to providing additional payments to LTCHs who have high percentages of end-stage renal disease (ESRD) patients because of the financial burden these patients have on an LTCH. In Section 307(b) (1) of the BIPA, there was a mandate that the Secretary shall examine, and may provide for, adjustments to payments under the LTCH PPS, including adjustments to DRG weights, wage adjustments, geographic reclassification, outliers, updates and disproportionate share adjustment. We believe that high utilization of dialysis in some LTCHs would require the need to create a new “outlier” pool for these facilities. In past years, RML suggested that LTCHs be provided the same dialysis add-on payment that is provided to short stay hospitals. RML is attaching as Exhibit A, a study that RML commissioned through KNG Health Consulting, LLC entitled “The Impact of Dialysis on the Cost of Care in Long Term Care Hospitals.” This study utilized the 2009 Medicare 100% Standard Analytic Inpatient File (SAIF) to identify all


discharges from LTCHs. This study concluded that dialysis patients receiving care in an LTCH are more costly than similar non-dialysis patients, yet there is no added payment from Medicare.

This study concluded that these increased costs for treating dialysis patients are not being properly accounted for under the current LTC-PPS. Accordingly, we requested that CMS amend their practice to either provide additional funding to LTCHs with high levels of dialysis patients or to modify and adjust the MS-DRG system to provide a CC or MCC for patients on dialysis. CMS has so far chosen to dismiss this proposition. RML Specialty Hospital’s percentage of Medicare patients requiring dialysis treatment has fluctuated between 17 to 23% in each of our last 5 fiscal years. During this time, our average cost to treat such dialysis patients exceeded our cost to treat similar non-dialysis patients. This cost disparity is the result of longer lengths of stay as well as higher patient costs associated with dialysis. As an example, our current contracted per-dialysis run cost is $415 which is above and beyond our routine daily costs. The KNG analysis concluded that the current payment system is insufficient to fully reimburse LTCHs for the increased costs associated with providing dialysis services. In fact, after controlling for LTC-MS-DRG assignment, high cost outlier status, short stay outlier status, and wage factors, KNG found that dialysis patients had standardized costs that were on average 11 to 12% higher than non-dialysis patients. Therefore, those few LTCHs who treat a disproportionately high number of Medicare dialysis patients are forced to operate at a significant cost disadvantage to other LTCHs under the current reimbursement rules. This situation causes undue financial hardship and jeopardizes our ability to continue providing such services in the future. As was also identified in the attached report, there are a relatively small number of LTCHs which have high numbers of dialysis patients. A Kennell/RTI report found that a higher percentage of Medicare beneficiaries receive ESRD, hemodialysis services in LTCHs than in any other provider setting. The percentage of Medicare beneficiaries who receive hemodialysis services in LTCHs is almost 200% higher than the percentage of Medicare beneficiaries who receive hemodialysis services in general acute hospitals and is the highest percentage of Medicare patients who receive hemodialysis services in all Medicare provider types.

Additionally, there are complexities associated with a patient that has a long stay in a short stay hospital and then transfers to the LTCH setting. Issues such as wounds, wound débridements, specialty beds, significant antibiotic treatments, etc. are not accounted for in the LTC-DRG system. We do not believe these higher complexity patients that come from short stay hospitals are fully incorporated into the proposed risk adjustment model. An additional “risk” that is not accounted for in the current model is for those patients who lack a safe and appropriate discharge destination in which to transfer the patient after their course of treatment is completed at the LTCH. As CMS is aware, there are different treatment practices among LTCHs on a national basis. Some LTCHs are able to admit very complex patients because there are discharge options available in their particular marketplace. Alternatively, there are LTCHs that are unwilling to take patients because there may be no viable discharge options. This needs to be incorporated into the risk adjustment methodology. Patients can be admitted in the Chicago area who are on


vent, dialysis, and may have significant wounds. However, this patient may never leave the short stay hospital in many other markets.

This metric may not be sensitive enough to pick up the significant variations that can be encountered within a specific MS-DRG. As an example, MS-DRG 207 relates to a patient on mechanical ventilation greater than 96 hours. This category might have patients with wounds, dialysis, might include a patient only going to the LTCH with the express purpose of weaning to nocturnal ventilation, etc. This is a significant problem as RML does not believe the normalization process for this metric will pick up these patient complexity differences. An LTCH like RML with over 500 ventilator discharges a year takes a much different type of patient than an LTCH that has few ventilator patients. As an example of this impact, there is no difference between what Medicare would spend for a patient that is discharged at day 32 vs. day 50 of their stay. Day 32 at the LTCH would be within the standard rate. At day 50, the patient may or may not have reached an outlier status. Still, there would be no additional payment received from Medicare, although the LTCH would continue to absorb added expenses until the patient reaches the high cost outlier category. Is the LTCH site neutral metric going to be adjusted for differences between the IPPS comparable rate for LTCHs compared to the IPPS system? How does Acumen plan on identifying and adjusting for LTCH site-neutral wound cases? The LTCH MSPB for wound cases will likely always be worse than the IPPS comparable metric because of the impact of a much higher length of stay in the LTCH.

TRIGGER EVENT It is identified that the associated services start at the trigger event for the LTCH and ends 30 days after the end of the treatment period. RML has been engaged in a number of post-discharge analyses and has obtained data from Medicare to study the pathways that patients take once they leave RML. This information has been invaluable and we have shared it on a number of times with representatives of CMS, Medpac and others. The amount of post-acute discharge variation that occurs in the LTCH patient population in Chicago is significant. Our internal studies suggest that a more appropriate time frame for the associated services would be 180 days. Although the magnitude (i.e. level of “spend”) of the metric would be much higher, we believe it is the appropriate time frame for a more complete and thorough recovery period for the types of patients we treat. In Table 1 - MSPB-PAC Episode Triggers, it is identified that a readmission within 7 days does not trigger a new episode but a readmission after 8 or more days triggers a new episode. We suggest that CMS utilize the existing 9 day LTCH interrupted stay definition that is currently in place. In Table 3 – Types of Service Categories Assessed for Exclusion, under paragraph 3, it identifies that a “group of clinicians” developed a list of service level exclusions using a framework which included: planned admissions, routine management of certain pre-existing chronic conditions, services related to inborn or congenital diseases, and some routine screening and healthcare maintenance. Please be aware that many of the patients we treat at RML are not “routine”. As an example, many patients may receive dialysis, but they may not yet be identified as reaching the ESRD level at the time of their admission. ESRD status designation may not occur until the


patient has been in the LTCH setting for some period of time. Please identify the process that will be utilized by CMS or its contractor to determine service exclusions in the future.

In Section 3.2.1 there is a review of the Exclusions from all MSPB-PAC Measures. We suggest that patients who were transferred from one short stay hospital to another short stay hospital prior to the admission to the LTCH be excluded from the LTCH measure. The rationale for this consideration is that these patients began their course of care at a community hospital and then were likely transferred to a regional referral or tertiary care center. We have found that these patients are typically sicker, have longer lengths of stay, have been on stronger courses of antibiotic therapy and generally have a higher complexity associated with their care and treatment requirements. In Section 3.2.3- MSPB-PAC Measure Calculations, the various steps to calculate the metric are outlined. There is a reference to a list of independent variables determining the 90 day look back window used in the risk adjustment model. This methodology again does not allow a provider to calculate its own metric and we will therefore be totally reliant on CMS for this metric calculation. A methodology is utilized for extreme predicted values which are truncated and then renormalized to maintain a consistent episode payment. Can the contractor or CMS provide information on the definition of an extreme predicted value? We believe an organization such as RML which often has 20% - 25% of our entire patient population identified as high cost outliers, would fall within the definition of an extreme predicted value. We are aware of no methodology which would allow a renormalization process to occur for this large of a group of patients. Would it be easier to eliminate these patients from the calculation entirely?

The MSPB-PAC measure is identified as an opportunity to assess a PAC provider’s resource use during an episode. Additional parameters that need to be included in the risk adjustment would be for things such as central lines, tube feedings, dialysis, wound management, etc. RML is very concerned that this metric (by itself) could lead providers to make determinations on the types of patients they are willing to admit. This metric should not be used as a determining factor in this decision making process. RML believes this measure is currently not a good indication of a provider’s ability to compare itself to others nor is it a metric associated with the quality of care provided. It is solely a payment indicator that has no relationship to outcome or the quality of services that are provided. I appreciate the opportunity to comment on the proposed Draft Measure.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 4, 5, 8, 13, 15, 17, 21, 22, 26, 29, 44, 48, and 49.




Name, Credentials, Organization, and Email of Commenter: Peggy Kirk, Senior Vice

President, Chief Clinical Operating Officer, Rehabilitation Institute of Chicago [email protected]

Type of Organization: IRF

TEXT OF COMMENT:

On behalf of the Rehabilitation Institute of Chicago (“RIC”), we appreciate the opportunity to comment on the Draft Specification for the Medicare Spending Per Beneficiary. We appreciate your attention to our comments, questions and recommendations, and we are available to assist you if we can provide any further information.

RIC operates a research-based health care system specializing in providing comprehensive rehabilitation services to the physically disabled through an array of diagnostic and therapeutic services. Its mission is rooted in its dedication to providing the highest quality patient care and outcomes through integrated research, scientific discovery, and education. As part of this system of care, RIC currently operates a 182-bed licensed inpatient rehabilitation facility (“IRF”) hospital and provides a wide scope of outpatient services from its primary location at 345 E. Superior Street in Chicago, Illinois as well as multiple additional locations through wholly-owned or other alliance structures with other hospital systems throughout Illinois and in northwest Indiana.

Over the years, RIC has earned an international reputation for excellence in patient care, medical research, and professional training. In 2015, for the twenty-fifth year in a row, RIC was ranked by U.S. News & World Report as the leading rehabilitation hospital in the United States. In fact, RIC is the only hospital in the country of any kind that has earned this ranking for twenty-five consecutive years. RIC serves patients from around the globe; during the past year, approximately 60,000 patients from all fifty states and nearly forty-five countries received care from RIC.

RIC is also the Northwestern Feinberg School of Medicine’s Department of Physical Medicine and Rehabilitation physiatry residency program, which is one of the largest and most sought after programs of its kind in the country. RIC has eight federally designated research programs, including designations as: a Rehabilitation Research & Training Center; a National Center for Medical Rehabilitation Research; the Midwest Regional Spinal Cord Injury Care System; a Rehabilitation Engineering Research Center dedicated to stroke research; the nation’s only Outcomes Rehabilitation Research & Training Center; a Rehabilitation Engineering Research Center for technologies for children with orthopedic disabilities; a Rehabilitation Engineering Research Center for manipulation and mobility technologies; and a Rehabilitation Engineering Research Center for computers and robots in therapy.

RIC understands that the Centers for Medicare & Medicaid Services (“CMS”) has contracted with Acumen, LLC to develop the Medicare Spending Per Beneficiary – Post Acute Care (“MSPB-PAC”) Resource Use Measures. The contract name is Calculating Episode-Based Costs from the Medicare Episode Grouper for Physician Feedback. The contract number is HHSM-500-2011-000121, and the task order number is HHSM-500-T0008. As part of its measure development process, CMS has asked for comments on these draft measures mandated by the Improving Post-Acute Care Medicare Transformation Act of 2014 (“IMPACT Act”). CMS has asked that stakeholders provide comments regarding the overall episode construction



methodology, exclusions, and the risk adjustment approach. (CMS Public Comment Page, at https://www.cms.gov/Medicare/Quality‐Initiatives‐Patient‐Assessment‐Instruments/MMS/CallforPublicComment.html#11 (last visited January 27, 2016).

We appreciate the opportunity to provide the following comments, which are directed towards the MSPB-PAC measure for IRFs.

Summary of Proposed Resource Use Measure

The MSPB-PAC resource use measure assesses average resource use of an IRF provider in comparison to other IRF providers. Resource use is determined during a period of care that the Draft Specification calls an “episode.” For a patient who is receiving care from an IRF, the episode starts on the day of IRF admission and continues for thirty days after discharge from the IRF.(Ibid at Sec. 3.1) The costs associated with an episode are from services the IRF provides as well as from providers other than the IRF. The costs incurred in an episode are all attributed to the IRF, unless excluded by the measure.

An episode is divided into a “treatment period” and an “associated services period”. The “treatment period” begins at admission to the IRF and ends at discharge from the IRF; all “services that are provided directly or reasonably managed by the attributed post-acute care (“PAC”) provider” would be included. (Ibid. at Sec.3.1) Certain services would be excluded from the treatment period and the associated services period if “they are clinically unrelated to PAC care and/or because PAC providers may have limited influence over certain Medicare services delivered by other providers during the episode window.” (Ibid. at Sec. 3.1.5)

RIC appreciates the intent of the measure and acknowledges and agrees with important considerations within it. Specifically, we appreciate that comparative costs are specific to each PAC setting; that a national median cost is used as a reference, rather than a national average cost; that “prior care status” is recognized as impacting needed services and is included in the risk adjustment methodology; that “day of admission” claims related to decisions made prior to admission are excluded; and that additional exclusion criteria are also considered.

As is set forth below, RIC’s comments are focused on exclusions, the overall episode construction methodology, and risk assessment. However, as an overriding comment, the proposed measure is intended to address a complicated and nuanced area of health care which is intended to be made available to patients and their families. Thus, it is critical that the measure be understandable and meaningful to the public. Moreover, patients and their families who will be reviewing the information CMS provides on the basis of the proposed measure are often at a fragile and difficult time in their lives. The complexity and nuances of the measure, combined with the position patients and their families find themselves in, increases the odds of misunderstanding and confusion among those who choose to examine comparative resource use across PAC settings.

In particular, we recommend that any public presentation of the measure’s information, such as on CMS’s website, be provided in connection with information on outcomes and quality, information that is key to beneficiaries as they select a PAC setting. While the Draft Specification suggests costs should be viewed in relation to quality, the Draft Specification does

https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/MMS/CallforPublicComment.html#11


not explicitly include outcomes and quality measures. RIC recommends that the resource use measure be linked with other quality and outcome measures, and be reported in that manner.

Finally, it is unclear how a comparative cost measure that includes inpatient, outpatient, part A and part B costs for multiple disparate episodes and diagnoses would be “actionable,” which is a stated goal of the proposal. RIC recommends CMS provide comparative costs within each of these categories in order to identify priorities and opportunities and thus make it actionable.

Comments on Exclusions

1. The final list of excluded services should be made available for public review and comment.

While the Draft Specification sets out a process for identifying services to be excluded, it does not identify most of the services that would be excluded under the proposed measure. (See ibid at Sec. 3.1.5) The Draft Specification states that “Certain services are excluded from the treatment and associated services periods because they are clinically unrelated to PAC care and/or because PAC providers may have limited influence over certain Medicare services delivered by other providers during the episode window.” (ibid. at Sec. 3.1.5) Chemotherapy and dialysis are listed as examples of such services, but the final list of excluded services is still under development using a process described by the Draft Specification. (See ibid. at 3.1.5 (“Lists of clinically assessed service exclusions…are being further refined by [CMS] clinicians as well as a group of independent clinicians…The service exclusion rules developed during this exercise will become part of the specifications for used for constructing MSPB-PAC episodes”)

We appreciate the explanation of the process being used to identify exclusions. However, the description of the process is insufficient to be able to comment on whether the final list of exclusions is appropriate or requires further review. Once a list of excluded services has been completed, that list should be made publicly available for review and further comment.

2. CMS should consider additional exclusions during the first day of an episode.

When admitting patients to an IRF from an acute care hospital, the admitting physician often will rely on the medical judgment and medical record of the acute care hospital discharging the patient. At times, the IRF provides care on the first day of services that results from care provided at the discharging hospital. For example, a patient may be admitted to an IRF in an unforeseen unstable condition requiring evaluation and treatment or acute care readmission, all of which is unrelated to their rehabilitation care.

The Draft Specification already provides for excluding certain treatment services occurring on the first day of an IRF episode, including ambulance transport to the attributed PAC provider facility and DMEPOS orders preceding the patient’s admission to the IRF.(ibid at 3.1.3) RIC believes additional exclusions for first day of care should be included. In particular, patients who are admitted and discharged on the first day should not be counted in the IRFs measure, either for the treatment window or the associated services periods of an episode.

Comments on the Overall Episode Construction Methodology


3. RIC has concerns regarding including all non-treatment services that occur within the associated services period.

As proposed, the purpose of including post-discharge costs of care appears to be to incentivize PAC providers to reduce the costs of post-discharge care providers. Specifically, the Draft Specification states that episodes to “overlap with hospital and other PAC episodes. Allowing for this overlap ensures that there is alignment of incentives between settings to ensure integrated, efficient care for any given beneficiary.” (ibid at Sec 3.1)

The unintended effect of this measure could result in some PAC providers recommending less post-discharge care and/or steer patients towards low-cost and low-quality post-discharge providers. This will make PAC providers who do not cut corners appear more expensive. Most importantly, it will negatively impact patient care unless it can be demonstrated that the lowest cost will provide the same outcomes. In fact, as mentioned above, there are no outcome measures explicitly associated with this measure at all, further unintentionally incentivizing providers to limit services during a patients stay through length of stay management or minimizing post discharge services.

Furthermore, many PAC providers do not currently have the means to reasonably manage post-discharge care of another provider. For example, a patient discharged from RIC may be discharged home with additional home health services. The patient, not RIC, selects which home health provider to use. Similarly, if a patient is discharged to a skilled nursing facility, the patient selects the SNF. In both instances, RIC does not have the ability to control or manage the care the patient receives from the HHA or the SNF.

We acknowledge that an IRF can make recommendations about post-discharge PAC care that increases beneficiary resource use. However, we have concerns about the inability to control the costs within those settings, in particular when patients select post-IRF SNFs and HHA. Additionally, in certain instances, a patient may select post-discharge PAC care that the IRF believes is not appropriate. For instance, a patient may select to be admitted to a SNF after discharge from an IRF, even when the IRF disagrees that the patient requires SNF care. The patient, or the patient’s legal representative, makes the decision and the SNF decides whether admission is appropriate. A physician order from the IRF is not required for admission.

Requiring IRF providers like RIC to manage post-discharge care becomes further exacerbated when they provide care to patients from around the country. These patients often wish to be discharged to their home, or if they need further PAC care to a PAC provider closer to home. Given the geographical distances involved, IRF physicians, nurses, or care managers cannot control post-discharge PAC care provided to, for example, a patient who leaves Chicago for SNF care closer to her home in Maryland, or Colorado, or California. The costs in the associated services period related to care beyond a certain distance from the IRF should be excluded from the measure.

Of particular note, including post-discharge care in the resource use measure favors large, national providers and so reducing the number of providers who are able to compete in a given market. Over time, the effect of such a measure may actually increase costs, rather than reduce them, as providers consolidate nationally in order to direct post-discharge care, and so reduce competition for providing health care services.


RIC supports integrated, efficient care when it results in achieving the highest potential outcome. However, reliance on a resource measure simply is not an appropriate mechanism to ensure beneficiaries have integrated, efficient care after discharge from a PAC provider.

Treatment Period

4. The readmission gap length of seven days should be revised to be consistent with the existing Two-Midnight Rule.

The IRF measure “treat[s] readmissions for the same patient and provider within 7 or fewer days as part of the same treatment period to reflect the likelihood that these closely adjacent stays are related.” (Draft Specification at Appendix D.1) A new episode for a patient who is readmitted after 8 or more days from discharge. According to the Draft Specification, the gap length was selected after conducting empirical analysis of Medicare FFS claims data. Additionally, the SNF and LTCH measures use a 7 day period and “TEP panelists and CMS clinicians were in favor of a consistent period across the institutional PAC settings.”

The Draft Specification does not include the results of the empirical analysis, making it difficult to comment on the quality of the analysis or the results. However, since each PAC settings measure is used to compare a PAC site’s resource use to other PAC sites in the same setting, it is not clear why a consistent period would be preferred for the measures for IRFs, LTCHs, and SNFs.

Instead, we suggest aligning the gap length for the IRF measure with existing regulations for IRFs in this area. Specifically, IRFs are subject to the Two-Midnight Rule, and IRFs already treat readmissions after three days or more as a new admission. Aligning the gap length with existing regulation will likely improve efficiencies in IRF’s understanding, implementing, and reporting of the measure.

Comments to Risk Adjustment

5. Additional risk adjustment measures, such as function and demographic factors, should be included.

The Draft Specification acknowledges that “the purpose of risk adjustment is to compensate for patient health circumstances and demographic factors that affect resource use but are beyond the influence of the attributed provider.” (ibid. at Sec. 3.2.1.) In order to ensure completeness, additional risk adjustment measures should be included to the proposed measure.

Many of our patients suffer serious, life-altering injuries: car accidents that result in spinal cord injuries, strokes that leave patients in a locked-in state, and other conditions that strip them of their functional abilities. At RIC, we give our patients hope that they can return to the activities that give meaning to their lives. Being able to increase a patient’s function is key to the work of RIC and other similar IRFs. It also is becoming increasingly understood in the research literature as a key indicator of hospital readmission rates. (See, e.g., Greyson et al, Functional impairment and hospital readmission in Medicare seniors. JAMA Intern. Med. 2015, available at http://www.ncbi.nlm.nih.gov/pubmed/25642907 (“Functional impairment is associated with increased risk of 30‐day all‐cause hospital readmission in Medicare seniors, especially those

http://www.ncbi.nlm.nih.gov/pubmed/25642907


admitted for heart failure, myocardial infarction, or pneumonia. Functional impairment may be an important but under addressed [sic] factor in preventing readmissions for Medicare seniors.”)

RIC strongly suggests including a patient’s functional ability as a risk adjustment factor. Doing so would be consistent with the hospital MSPB measure, which adjusts for “disability.” (Draft Specification at Sec. 3.2.2) Function information is already available in CWG information that IRFs already provide, which means that CMS already has access to this information.

Additionally, race and economic advantage often play major roles in a patient’s recovery. For example, patients who are economically advantaged have better support networks or other means to provide for post-discharge care. Adding demographic factors to the risk adjustment measures is also consistent with the hospital MSPB measure, which includes demographic factors. (ibid)

6. RIC recommends reconsideration of the six clinical case-mix categories

The Draft Specification includes six clinical case-mix categories of prior care. Under the measure, an episode would be placed into one of the six categories. When multiple claims have the same end date, priority is given to the categories in the following order: Prior Acute Surgical IP – Orthopedic; Prior Acute Surgical IP – Non-Orthopedic; Prior Acute Medical IP with ICU; Prior Acute Medical IP without ICU; Prior PAC; and Community. The priority list was determined as follows: “Procedures are at the top of the hierarchy as they are typically easier to evaluate, with the remaining categories ranked in decreasing severity.” Claims are then risk-adjusted only with other claims in the same category. (ibid)

RIC agrees with the Draft Specification that “prior care status has an effect on the level of care needed while in PAC….” (ibid) However, the patients who receive care at RIC do not fit neatly into the categories as identified, because of the quaternary group of patients seen. For example, a spinal cord injury patient will have had surgery and ICU care at the prior setting of the acute care hospital. That patient, however, would be classified as “Prior Acute Surgical IP – Orthopedic.” The categorization does not take into account the resources that a patient with prior surgical and ICU care requires.

Other examples are readily available:

• Patients who were in ICU care for a few days at the prior acute care hospital often require fewer IRF resources than do patients who are in the ICU for weeks or longer.

• Patients who are admitted from an LTCH will generally require more resources than a patient admitted from a SNF, who will generally require more resources than a patient admitted from an HHA – yet all these patients would be categorized into the “Prior PAC” category.

• Patients who had surgery for a spinal cord or brain injury would be categorized as “Prior Acute Surgical IP – Non-Orthopedic” yet have substantial resource requirements while in IRF care.

RIC recommends reconsideration of whether the categorization approach appropriately risks adjusts for prior care status. At the least, additional risk adjustment indicators should be added for the examples given above.


7. The full risk adjustment model, including risk adjustment coefficients, should be made available as soon as possible for full review.

The Draft Specification lists the covariates under consideration for risk adjustment, but does not provide the full risk adjustment model, including the weighting of each covariant. As a result, we cannot comment on whether the model appropriately weighs the identified covariates. The full model should be made available to the public for review and comment.

Conclusion

We support the IMPACT Act requirements related to the prompt development and implementation of this measure and others, and appreciate CMS’s efforts in making this measure available for review and comment.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 4, 5, 8, 11, 13, 17, 24, 26, 28, 30, 31, 35, 40, 41, and 48.

Regarding your comment on first day exclusions, an acute inpatient hospitalization with an admission and discharge on the first day of an IRF episode is not counted toward the IRF episode, as specified on Appendix B of the draft measure specifications document in this paragraph: "Firstly, the claim representing the transfer source is removed if it has a discharge date occurring on the first day of the episode. The transfer source may be an IP, SNF, IRF, LTCH, or HHA claim. For example, for a patient transferring from IP to a SNF on the first day, the IP claim is not counted toward the SNF episode. This exclusion of claims associated with the transfer facility occurring on the first day of a PAC episode applies to all settings."

Regarding your comment on favoring large, national providers, this makes an assumption about the effect of the measure that is unverified, as small providers may be very capable of providing high quality care that leads to low post-discharge resource use, and hence low overall resource use. The predicted result has not occurred with the use of the MSPB measure in acute care hospitals, and we do not have evidence it will occur in PAC providers.



Name, Credentials, Organization, and Email of Commenter: Bruce M. Gans, MD, Chair,

AMRPA Board of Directors, Executive Vice President and Chief Medical Officer, Kessler

Institute for Rehabilitation, National Medical Director for Rehabilitation, Select Medical

[email protected]


TEXT OF COMMENT:



The American Medical Rehabilitation Providers Association (AMRPA) is pleased to submit these comments on the draft specifications for the MSPB-PAC measure as applicable to inpatient rehabilitation facilities (IRFs). We recognize that this measure is required under the Improving Medicare Post-Acute Care Transformation (IMPACT) Act. As a result, our comments are designed to suggest improvements to the measure to ensure it achieves its intended goal of measuring resource use in a way that protects access to these services for Medicare beneficiaries and does not impose an undue burden on PAC providers.

AMRPA is the national voluntary trade association representing more than 500 freestanding rehabilitation hospitals, rehabilitation units of general hospitals, and outpatient rehabilitation service providers. Our members provide medical rehabilitation services in a vast array of health care settings working with patients to maximize their health, functional skills, independence, and participation in society so they are able to live as independently as possible at home, return to work or, in many instances, pursue an active retirement. On average Medicare Part A payments represent more than 60 percent of IRFs’ revenues. (MedPAC, Report to Congress 240 (Mar. 2015)

Section 2.5 Brief Description of Measures

AMRPA is pleased to see that a separate MSPB measure are being developed for each PAC setting and thus comparison will be drawn between providers of the same type as opposed to trying to draw comparisons among different types of PAC providers. AMRPA strongly agrees that “setting specific measures allow for more meaningful comparisons to be made between providers than if one single measure were calculated across all providers in all PAC settings,” (page 5). Furthermore we encourage the Centers for Medicare and Medicaid Services (CMS) to continue its development of cross-setting PAC measures with this approach. We recommend Acumen and CMS adopt this approach in any final definition of the measure. We support the decision to use the national median based on the prospective payment system (PPS) base payment as the comparison benchmark, which does not include add-on payments or facility adjustors.

Section 2.5.1 Numerator

The numerator description includes several terms that need to be defined before we can provide complete comments. They are: “standardized episode spending,” “expected episode spending,” and “average standardized episode spending across all providers.”

Section 2.5.2 Denominator

The denominator is defined to specify that the “Measure is the weighted median MSPB-PAC Amount across all episodes for IRFs nationally,” but insufficient information is provided about the weighting methodology.

Section 3.1 Episode Construction

The draft measure specifications describe an episode window comprised of two parts: the treatment period, and the associated services period. The treatment period includes the time period from admission (trigger event) to discharge from the PAC provider and the associated


services period includes the services provided during the PAC stay and within 30 days of discharge from the PAC provider, with certain exclusions.

Overall we support the inclusion of services provided during the PAC stay and for 30 days post-discharge in the episode window. However, we are concerned that the proposed two-part composition of the episode (treatment period and associated services period) and their respectively attributed services creates an MSPB measure(s) that is unnecessarily complicated. Although the draft measure specifications state this bifurcation is necessary “because clinical exclusions of services in the treatment period may differ from clinical exclusion of services in the associated services period,” it does not provide any demonstrative examples. In the absence of further information, AMRPA recommends that those services excluded from the MSPB episode during the treatment window, such as chemotherapy and other routine maintenance services, and planned admissions, also be excluded from counting towards the MSPB episode if they occur during the associated services window.

The draft specifications state, “The definition of PAC episodes allows episodes to overlap with hospital and other PAC episodes,” (page 11) and “the full payment for all claims that begin within the episode window is counted toward the episode” (page 38). If we understand these statements correctly, one beneficiary’s trajectory of care could initiate multiple PAC MSPB episodes across different settings. For example, a patient who is discharged from an IRF and at some point during the associated services period is admitted to another PAC provider such as a SNF would trigger multiple episodes. In that scenario, the spending from the SNF stay will trigger a new MSPB episode for the SNF (due to SNF admission) and will also be attributed to the discharging IRF’s associated services period. In other words, the Medicare expenditures for a stay at one PAC setting will be associated with two MSPB-PAC episodes. This form of attribution is not inherently problematic and AMRPA supports this construction of the MSPB-PAC measures as long as the measure remains a comparison of providers of the same type (e.g., IRFs to IRFs) as currently proposed. If, however, at some point in the future CMS intends to compare different type of providers (e.g., SNFs to IRFs), this type of attribution would be inappropriate because it “doubly” ascribes one Medicare expenditure (the SNF stay in the example above) to two PAC providers. We request that Acumen and CMS clarify that this type of attribution will only occur while comparisons are taking place between PAC providers of the same type as opposed to across PAC providers.

Section 3.1.5 Step 5: Excluding Clinically Unrelated Services

The draft measure specifications note that certain services are excluded from the episode window (either the treatment period or associated services period) because they are clinically unrelated to PAC care and/or because PAC providers may have limited influence over certain Medicare services delivered by other providers. We agree that certain exclusions are appropriate and necessary to avoid unintended consequences as outlined in the draft document. Therefore, we recommend that costly and routine care not related to the IRF stay, such as dialysis or chemotherapy, be excluded from the IRF MSPB episode.

Several examples of excluded services are provided throughout the draft measure specifications including on Table 3 (page 16) and in Appendix A. However, the document states while lists of clinically assessed service exclusions have been developed for each PAC setting by clinicians from CMS and the measure development contractor, these lists are being further refined by these


clinicians as well as an independent group of clinicians. Without more definitive information on the exclusions under consideration, we are limited in our ability to provide constructive feedback on whether such exclusions are appropriate. AMRPA supports the exclusions specified in Appendix A, Table A-4 and recommends CMS adopt them and requests the opportunity for the field to review any additional refinements before they are finalized.

Section 3.2.1 Implementing Episode-Level Exclusions

AMRPA supports the four beneficiary exclusions listed in the draft specifications for beneficiaries enrolled in Medicare Advantage, beneficiaries not continuously enrolled in fee-for-service (FFS) for the lookback period, beneficiaries whose primary payer is not Medicare, and beneficiaries who die during the episode. In addition, patients who are discharged against medical advice (AMA) should also be excluded because a PAC provider is unable to manage those patients’ downstream trajectory of care and associated costs.

Section 3.2.2 Risk Adjustment Approach

Among the risk-adjustment factors CMS and Acumen intend to use are 70 Hierarchical Condition Categories (HCCs). AMRPA believes risk adjustment for patient severity for IRFs is more appropriately accomplished by using case mix groups (CMGs) rather than HCCs because HCCs were originally developed for risk-adjustment in the Medicare Advantage program and are based on ICD9-CM codes. Accordingly, CMS and Acumen should use the CMGs rather than HCCs for IRFs.

In addition, we believe the measure should be risk adjusted for socioeconomic and sociodemographic status (SES/SDS) as outlined in the National Quality Forum (NQF) report entitled Risk Adjustment for Socioeconomic Status or Other Sociodemographic Factors: Technical Report. These characteristics would include factors such as the availability of caregiver and community supports and race. Short of incorporating a fleshed-out SES/SDS risk-adjustor, a beneficiary’s dual eligible status could be captured from claims data as a proxy for socioeconomic/demographic status.

Section 3.2.3 MSPB Measure Calculation

We appreciate that the measure calculation is relatively complex in part to ensure the fairness of the calculation. Nonetheless, we believe it is important that IRFs be able to replicate the calculation and verify the MSPB value assigned to them. As currently constructed, the calculation is too complex to enable them to do so. We also believe that a single value indicating the average spending provides little actionable data which PAC providers such as IRFs can utilize to manage spending more effectively. We believe it is CMS’ intention to be transparent, which is not achieved by the measure’s calculation as currently constructed.

The draft measure specifications also do not delineate if or how a PAC provider who disagrees with the value assigned to them through the measure calculation may appeal the determination. It is critically important that such an appeal mechanism be available to providers.

Finally, through a post-payment audit the provider may be required to return funds to CMS that were part of the MSPB calculation. If such an event occurs, the provider’s MSPB value should be recalculated to reflect the lower spending.


Appendix D, D.1 Collapsing Proximate Stays

AMRPA is concerned with the treatment of readmissions within a seven day window of discharge as outlined in this section. Specifically, we are concerned that the readmission counts towards the original MSPB episode only if the patient is readmitted to the same IRF. Instead, we believe that the readmission should be attributed to the original IRF regardless of whether the patient returns to the original IRF or another IRF. In fact, we believe this form of attribution should occur across PAC providers so that the PAC MSPB measure does not unintentionally incentivize provider behaviors that are not in beneficiaries’ best interests.

We are also concerned about setting a threshold, in this case seven days, as we believe it could lead to unintended consequences and efforts to manipulate the length of stay in a way that avoids attribution to the discharging provider. We encourage Acumen and CMS to reevaluate whether such as threshold is appropriate or necessary.

Conclusion

In closing, AMRPA appreciates the opportunity to provide our recommendations for improving the structure of the MSPB measure.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-4, 5, 8, 11, 14, 15, 16, 17, 25, 26, 29, 30, 37, 46, and 48.


Measure Set or Measure: Measure set

Name, Credentials, Organization, and Email of Commenter: Jayne Hart Chambers, Senior

Vice President Quality, Federation of American Hospitals [email protected]


TEXT OF COMMENT:

The Federation of American Hospitals (FAH) is the national representative of more than 1,000 investor-owned or managed community hospitals and health systems throughout the United States. Our members include teaching and non-teaching, short-stay acute, inpatient rehabilitation, long-term acute care, psychiatric and cancer hospitals in urban and rural America, and provide a wide range of acute, post-acute and ambulatory services. The FAH is pleased to have the opportunity to provide our comments on the draft Specifications for the MSPB-PAC Resource Use Measures.

The FAH recognizes that Acumen has been tasked within a very short period of time to create a complicated measure of resource use required by the Improving Medicare Post-Acute Care Transformation (IMPACT) Act. However, our members find it extremely challenging to comment in detail on the proposal in the two-week window of time (even with the three day



extension), and FAH encourages Acumen to grant additional time for future review of measures of such significance.

The FAH comments are intended to suggest improvements to the proposed measures to ensure they achieve the intended goal of measuring resource use while also protecting access to post-acute services for the Medicare beneficiaries and also not imposing undue burden on post-acute care (PAC) providers. The FAH realizes the MSPBPAC proposal is the first step in a multi-step process and strongly encourages Acumen to seek additional comments on draft MSPB-PAC resource use measures when detailed measure specification for the numerator and denominator are available.

The FAH supports the proposed development of a separate MSPB-PAC resource use measure for each of the PAC settings. The MSPB-PAC measures should NOT be used for cross-setting comparisons, particularly since there is not an appropriate methodology to standardize costs or payment structures across PAC settings. The FAH agrees with the report’s statement that “setting specific measures allow for more meaningful comparisons to be made between providers than if one single measure were calculated across all providers in all PAC settings,” (page 5). The “setting specific” methodology would be beneficial as Acumen and CMS develop and adopt additional measures under the IMPACT Act.

However, since all other IMPACT measures are cross-setting measures, it likely could be assumed that end-users may attempt to make cross-setting MSPB comparisons as well. Since Acumen has realized that such a cross-setting MSPB measure is infeasible, CMS and Acumen should actively warn doctors, hospitals, patients and families against using the MSPB-PAC resource use measures on a cross-setting basis.

The MSPB-PAC proposal introduces a number of new key terms that would benefit from more specific definition or a glossary of terms which the user can easily reference. For instance, the definition of an “episode trigger” for Skilled Nursing Facilities (SNF), Long Term Care Hospitals (LTCH) and Inpatient Rehabilitation Facilities (IRF) is different than it is for a Home Health Agency (HHA). The “episodes” for each of these settings are also different. Therefore, having a readily assessable resource to which to refer would be helpful to the reader. It also would be beneficial to have more robust definitions and examples of what services would be included in an episode for each setting.

The measurement period for each PAC setting is likely to overlap for a significant number of cases. Therefore, the definition of the episode and which services are included or excluded needs to be very explicit. Such specificity is critical to a facility understanding what costs are being attributed to it. The FAH appreciates that the measures propose to average the services and not sum the services across episodes. However, having a detailed example of how this would be implemented would be helpful for better understanding of the proposed model.

In addition, the FAH suggests that further detailed definition of terms used to describe the numerator, denominator and episode construction would inform our members’ understanding of the model. For instance, the “standardized episode spending,” “expected episode spending,” and the “average standardized episode spending across all providers” are not readily understood terms without further definition. For instance, in the denominator, the phrase “Measure is the weighted median MSPB-PAC amount across all episodes for IRFs nationally” does not define


the “weighting,” which is key to the definition. The FAH also would appreciate further explanation of the “clinical exclusions of services” during a treatment period or the associated service period.

The proposal notes that certain unrelated services are excluded from the episode window because they are clinically unrelated to PAC care or because the PAC providers may have limited influence over certain Medicare services delivered by other providers. The FAH supports this exclusion of certain services. While the proposal provides several examples on Table 3 and in Appendix A, it also indicates that clinicians are continuing to work on the development of the list of excluded services. Without thorough understanding of the exclusions under consideration, the FAH has limited ability to provide constructive feedback on the impact of the exclusions or the appropriateness of the services that have been chosen.

The proposal discusses risk adjustment and the intent to use 70 Hierarchical Condition Categories (HCCs). In the IRF setting, the FAH believes that the case mix groups (CMGs) would provide more accurate risk adjustment.

The FAH also strongly suggests including risk adjustment for socio-demographic status (SDS). Community resources play a significant role in both the overall costs of care and in a patient’s ability to recover. A growing body of evidence indicates that individual and community-level social factors have a significant impact on acute care hospital readmission rates. The National Quality Forum (NQF) report titled Risk Adjustment for Socioeconomic Status or Other Sociodemographic Factors: Technical Report also provides support for SDS adjustment. The FAH long has advocated for appropriate adjustment of SDS factors, and we believe this growing body of research has relevance for the post-acute settings resource use as well as the acute care setting.

The MSPB-PAC proposal does not discuss how the proposed MSPB-PAC measure might be affected by facilities that are included in one of the alternative payment model programs such as the (Comprehensive Care for Joint Replacement Model). We strongly encourage Acumen to address possible implications of utilization and cost variation when facilities active in an alternative payment model are being compared to facilities not in such a model. How will the proposed MSPB-PAC model be affected?

The FAH members are very concerned that the MSPB-PAC resource measure is constructed in such a way that they will not have full access to the data needed to replicate the measure on their own. A MSPB measure intended to facilitate better understanding of episode costs is beneficial only if a facility understands its own costs and can use that information for real-time improvement. FAH members want to be able to replicate for themselves the data presented at a national level. In the current model, the costs attributed to a facility would not be able to be replicated by the facility, and, as such, self-assessment and improvement would not be facilitated.

The complexity of the measure calculation raises concerns for FAH members. While the complexity of the measure calculation is partly due to the desire to ensure fairness, the complexity makes it very challenging for our members to calculate their own scores and to ensure they are assigned the appropriate MSPB score. A single value indicating the average spending provides little actionable data for the facility to manage effectively its own spending.


The FAH questions the transparency of the measure score as it is currently proposed. The FAH also recommends that an appeals process be developed so a facility can appeal a score if it believes there is a problem.

The proposal outlines the treatment of readmissions within a seven day window of discharge. The FAH is concerned that the readmission could count toward the originating MSPB episode only if the patient is readmitted to the same originating facility. Given that a number of facilities are included in the measurement process, the FAH recommends that the measure clarify that the readmission be attributed to the original facility regardless of whether the patient returns to the original facility or another facility. By developing this form of attribution across all PAC providers, the MSPB-PAC measure would make all such providers more accountable and avoid unintentionally incentivizing provider behavior that is not in the beneficiaries’ best interest.

The FAH also encourages Acumen to re-evaluate the proposed seven day window threshold. The FAH is concerned that such a threshold could lead to unintended consequences and potential incentives to manage the length of stay in ways that avoid attribution to the discharging provider.

Finally, the FAH strongly urges Acumen to conduct robust testing of the measure and to share widely the results of the testing so that all interested parties have the opportunity to better understand the impact of the measure.

Thank you for the opportunity to comment on the proposed MSPB-PAC Resource Use Measure.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-4, 5, 9, 11, 12, 13, 14, 15, 16, 17, 18, 25, 26, 30, 37, and 48.



Name, Credentials, Organization, and Email of Commenter: Dale N. Schumacher, MD,

MPH, President, Rockburn Institute [email protected]


TEXT OF COMMENT:

Thank you for the opportunity to comment on the MSPB Post-Acute Care January 2016 Acumen white paper.

Our experience with the inpatient MSPB measure has been extensive (See attached publication – “To Increase efficiency – Decode Medicare Spending Per Beneficiary”). Subsequently we have refined several of these measures particularly those relating to care beyond the index hospitalization.

Comments:



• MSPB entry criteria and exclusions – In the original 2013 MSPB distribution included large numbers of cases beyond the index admission, for example, Medicare Advantage. Having access to excluded cases is often helpful as we track the patients over time and place. Recommendation: Do not unnecessarily hide or limit non-index admissions in either in-patient or PAC.

• Ambulance deletion - We are appreciative of the exclusion of the ambulance as part of the Part B services. The dollar amounts expended for ambulance services are often enormous and distort Part B. Recommendation: While we encourage the exclusion out of Part B Services, a separate ambulance category would be useful – we can better understand the transfer processes.

• Part B physicians – Recommendation: Consider expanding the number of Part B physicians that are tracked and provide the claims amount for each rather than rank ordering.

• Dual Eligible – We have been unable to reliably map dual eligibles back to our MSPB data even considering the disabled categories. Recommendation: Develop a flag that indicates dual eligible either in annual issuances or even retrospectively.

• Readmissions – It is possible, with effort, to identify readmissions to the same hospital. Recommendation: It would particularly useful to identify readmissions and even better those associated with the CMS Readmission Reduction Program.

• CMS Certification Number (CCN) – Since hospitals within systems often use the same CCN it is difficult to track patients who are discharged to PAC and then are readmitted to the same CCN. Recommendation: Further define the Episode of Care by identifying the readmitting hospital.

• SNF/HH – It is difficult to track patients in and out of SNF/HH. Recommendation: Be clear regarding sequences and time of service.

• Episode of Care – With hospitals it is essential that we can “walk back” Post-Acute Care. Below at numbers 7, 8 and 9 include examples of episodes of care extending outside the inpatient stay. Recommendation: Make PAC data results transparent and accessible so that linkage across the continuum can occur. [The commenter provided an image of a flowchart here which is contained in the separate attachment to this public comment summary report containing a table of these verbatim comments titled “Medicare Spending Per Beneficiary – Post-Acute Care Measures: Public Comment Verbatim Report”.]

• Maryland hospitals are excluded from the MSPB process because of the waiver. Since

Part A and Part B claims data are available for Maryland hospitals we Recommendation that you provide the MSPB inpatient data to Maryland hospitals as a dry run. [sic] The initiation of the IMPACT Act will not be delayed in Maryland if hospitals have experienced working with MSPB.

• Non-Index Own Episode Exclusions – Some inpatient visits are excluded from being their own episode because they are “Non-Index Own Episode.” While we can usually find the episode it is associated with, it requires investigating each event. Recommendation: It would be very helpful for CMS to include the episode number to which the excluded visit is assigned.


• Inpatient reporting is listed by highest claims paid - Recommendation: We should also be able to access inpatient chronologically.

• Regression models - In the hospital MSPB models Major Diagnostic Categories are the base for regression models. Recommendation: It is not clear in the January 2016 white paper regarding the diagnoses, disability levels, RUGS, etc. that would be the base. Please clarify and references regarding the stability of such models would be appreciated.

[attachment: Schumacher, D., Felgner, L., Dobkin, E., Nerhood, F., Paroski, M., “to increase efficiency, decode Medicare spending per beneficiary” healthcare financial management association (June 2015) [sic]]


Regarding your comment on Maryland hospitals, since CMS will calculate the measure using administrative claims data, and currently no PAC providers in the U.S. have experience working with MSPB-PAC measures, we do not believe it is necessary to provide data in advance to Maryland hospitals.

Regarding your comment on regression models, in the hospital MSPB measure, regression models are estimated separately within each Major Diagnostic Category. The analogous approach proposed in the January 2016 documentation involves estimating separate regression models within each clinical case mix category. Based on changes to the risk adjustment model proposed in TEP feedback and public comments, we have revised our risk adjustment approach. The results, illustrating the magnitude and statistical significance of coefficients, appear in the accompanying documentation.



Name, Credentials, Organization, and Email of Commenter: Rev. Daniel P. Clark, RN,

Individual [email protected]


TEXT OF COMMENT:

I wish I had the time to give the IMPACT Act the attention it deserves. I am in favor IN PRINCIPLE of the IMPACT Act. The available information and proposals are intense and need a great deal of thought–which requires time I don’t have. However, I do want to address a couple of pressing issues.



1) I am concerned that the tools/assessments that are developed will eventually be used to compare assessments and risk adjustments between Hospital discharge assessments and the initial assessment by the Post-Acute Care provider (PAC).

I am an RN with many years of experience in Home Health. I have not done a formal study, but I have noticed a pattern in people who are discharged from hospitals and quickly started on HH. Even those who are started within 12-24 hours already have decreased strength, stamina, balance, and self-care abilities. I have personal experience with someone whom I helped at the hospital in the morning and then was with them the rest of the day. He received PT the morning of his discharge. That evening, he was still exhausted from the trip home (same city). He required more assistance to the bathroom. He could not dress himself. He required contact assistance with all ambulation (for balance) with a walker. I had to supply help with all transfers. This was a huge set-back from when I was with him that morning. When we started HH the next day, he was still behind where he had been. I have spoken to many other HH nurses, and this is a general rule. Physical Therapists who have treated someone in a hospital setting and then treat them in the home are usually surprised by this. I think CMS would even acknowledge this circumstance–which is why we MUST start people within 48 hours of DC. Even semi-objective measurements (e.g., TUG scores, Tinetti balance scores) will be worse. It’s just our nature. I don’t know how to account for this with IMPACT related assessments, but it needs to be considered.

2) I am concerned that none of the Risk Adjustment Factors include Financial Resources. Throughout my nursing career, I have worked in a Hospital, Nursing Home (NH), and (primarily) Home Health (HH). I lack the empirical evidence, but I can promise you that Home Health patients are the most affected by financial factors.

As an exaggerated example, consider the following: A billionaire and a homeless man (born on the same day, same nationality) both go into cardiac arrest. They are taken to the same Heart Specialty Hospital because it is the closest facility. The staff in the cardiac ER are going to fight equally to save their lives. Both receive angiocaths (coincidentally at the same location in the heart). The damage to the heart is identical. Their course of treatment in the hospital will be very similar, because both have Medicare. There may be a few differences in the hospital outcomes because of their mindset is a bit different. But, the outcomes will be similar. They are discharged on the same day. The billionaire will go back to his mansion, have easy access to new meds, have easy access to the best follow-up physicians, have many servants or contractors who can assist him with whatever he needs. If he needs to install exercise equipment–no problem. He can have a dietician modify his diet. His cook will do whatever is needed. Anything his Dr says, he can do. The HH nurses and therapists will have easy treatments.

The homeless man? He will struggle with everything. HH can visit him at the homeless shelter. But, he’s going to depend on Medicaid for medicines. Hopefully, his hospital Dr prescribed something that Medicaid covers. He’ll still need his copay. Usually, shelters will find a way to get him to a Pharmacy. Exercise? Diet? We can educate, but it’s tough to get those under control. Access to a regular Cardiologist? Probably not. Medication compliance is HORRIBLE!!


I know this is an exaggerated example, but I use it to demonstrate the very real world issues that our poor face. They require the most time and resources from a HH perspective. Our Medicaid only patients are the lowest paying patients (from a financial basis), yet they are the neediest and therefore the costliest. (That’s why so many agencies have stopped accepting Medicaid). The poor patients are the ones we end up having to do extra errands. We do many extra referrals and provide help to them that just isn’t needed for upper income patients. Even if they are not homeless, the low income status still has major effect when they have to choose whether to pay rent, buy food, pay utilities, buy meds, etc.

If that person goes to a NH, most of those needs disappear. He will have access to the right diet, exercise regimen, Dr, etc. Medication compliance is very controlled. Exposure to the elements is limited. Overall, the “poor” factors are not as relevant. They may not go away completely, but they are highly mediated.

An Outpatient clinic would be affected, but likely wouldn’t be used by a homeless patient. Some of my friends in that setting might argue with me on that.

I remember (to some degree) the history of Financial Factors and OASIS. OASIS thru OASIS B1 had Item Number M0160 Financial Factors on the Start and ROC OASIS. See pages 66-67 of https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HomeHealthQualityInits/downloads/hhqioasisvolume4.pdf (when printed out, it is pp 2.52 thru 2.53). I understand the problems with the question. It was too subjective. Poor reliability. Some thought it was too sensitive a subject (which is crazy). But the biggest problems was that IT WAS NOT REPORTED TO CMS (See # 2, Item clarification on pg 66). I remember asking one of the developers of OASIS on a national provider call why the question was dropped for the next OASIS, and he said it was NEVER involved in any of the equations. That was ludicrous. Everyone in our room was shocked. It could have been made more objective. It definitely affects outcomes, so it should have been risk adjusted.

One way that this information can be made more objective is to ask what other programs the person is eligible for or is utilizing. I’m going to list the original question and it’s answers (numbers 0-5). Then, I’ll list some more objective ways of obtaining similar information:

(M0160) Financial Factors limiting the ability of the patient/family to meet basic health needs: (Mark all that apply.) …

0 – None …

1 – Unable to afford medicine or medical supplies …

2– Unable to afford medical expenses that are not covered by insurance/Medicare (e.g., copayments) …

3– Unable to afford rent/utility bills …

4– Unable to afford food …

5 – Other (specify)

https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HomeHealthQualityInits/downloads/hhqioasisvolume4.pdf


M0XXX Financial Factors may limit the ability of the patient/family to meet basic health needs. These needs may be met using Governmental or Formal community based programs. Consider the categories below and mark if patient or household member(s) is eligible for or receiving government or formal assistance from any governmental or Formal community based program. This should not include family or friends that help occasionally or regularly. The program in parenthesis ( ) are examples only and are not inclusive of all possibilities: (Mark all that apply.)

6 – Food assistance (e.g., SNAP Food Stamps, Regional Food Banks, Meals-on-wheels, Salvation Army, Churches, etc.)

0 = No

1 = Yes

2= Eligible, but not receiving help

7 – Utility Services assistance (e.g., utility grants, TANF, Salvation Army, etc.)

[scored the same as # 6]

8 - Assistance with paying Rent or home repair (e.g., Section 8, HUD, I, VA, etc.)


9 – Assistance with purchasing medications (e.g., Medicaid, Pharmaceutical company based programs, etc.)


I threw this together really quickly, so I understand there are issues with these. But, any Post Acute Care Risk Adjustment should include financial factors of some kind.

Sorry for the length, but thank you for your consideration.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment ID-30.

Regarding your comment on patient discharge, the IMPACT Act-related assessments ultimately are geared to encourage coordination of care for the patient and to address dilemmas such as the one you described. It is a responsibility of the hospital to ensure that patients are ready for discharge and coordination of care is a shared responsibility. The patient should receive referral to the adequate providers, and then when discharged, it should be coordinated that the patient is to receive post-acute care in a timely manner. We agree with your description of a current challenge and hope that this measure will encourage the solution you have proposed.




Name, Credentials, Organization, and Email of Commenter: Kate Jones, MSN, RN, CCM,

Senior Vice President and Chief Clinical Officer, Amedisys, Inc. [email protected]

Type of Organization: HHA and hospice organization

TEXT OF COMMENT:

Amedisys, Inc. (“Amedisys”), a national home health agency (“HHA”) and hospice provider providing care in 34 states through more than 450 Medicare-certified home health and hospice centers, appreciates the opportunity to comment on the draft specifications for the Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) resource use measures just released on January 13, 2016. However, given the complexity of the proposed measures and the significantly compressed comment period, we cannot provide substantive feedback at this time. We strongly urge Acumen and the Centers for Medicare and Medicaid Services (“CMS”) to reevaluate the comment period and provide a more realistic timeframe for interested parties to review the proposal and provide thoughtful, thorough, and support responses. RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment ID-9.

COMMENT 27 OF 45 Date: 1/29/2016 and 2/5/2016


Name, Credentials, Organization, and Email of Commenter: Mary K. Carr, V.P. for

Regulatory Affairs, National Association for Home Care & Hospice [email protected]

Addendum. Theresa M. Forster, VP for Hospice Policy & Programs, National Association for

Home Care & Hospice [email protected]


TEXT OF COMMENT:

The National Association for Home Care & Hospice (NAHC) is the nation’s largest trade association representing home health and hospice agencies including Visiting Nurse Associations, government-based agencies, multi-state corporate organizations, health system affiliated providers, and freestanding proprietary agencies. NAHC members serve over 3 million Medicare home health and hospice beneficiaries each year.

NAHC appreciates the opportunity to comment on the draft MSPB-PAC measure specifications. We wish to offer the following comments and recommendations.

Episode Construction and Window





• The proposed treatment episode for home health patients begins on the first day of the home health claim and ends 60 days after for all episodes except partial episode payments (PEPs). NAHC has concerns regarding the construct of the episode window and episode length for home health patients that are not under a home health plan of care for the entire 60 days (treatment episode). Patients are often discharged from care prior to the end of the 60 day home health episode. Since the treatment period is for the full 60 days, these patients could be under a MSPB treatment episode for days, even weeks, where care is not being provided by the agency. In addition, since the associated episode does not end until 30 days after the treatment episode ends, MSPB could include costs that occur months after patients have been discharged from the home health agency’s care. The proposed episode window unfairly attributes spending for home health patients discharged prior to the end of the 60 day episode.

NAHC recommends the treatment episode end at discharge for home health patients discharged prior to the end of the 60 day episode with the associated episode extending 30 days after the discharge date.

• Since each home health claim triggers a separate treatment episode it is unclear how costs will be attributed to the episodes when consecutive home health episodes occur. For example, the patient will be in the associated period for the first episode and in a treatment episode for the second episode.

NAHC is concerned that the home health episodic payment will be attributed to both MSPB-PAC episodes because a home health claim for a consecutive episode will have episode start date that falls within the associated period of the previous home health episode. NAHC requests the developers assure that costs will not be counted for both home health episodes.

Denominator

• The developers propose to use the national median for the MSPB-PAC Amounts for the denominator across providers in the same setting. Due to regional difference in costs for health care service and variations in practice patterns, using provider-specific medians or regional median for the MSPB–PAC Amounts seems more appropriate.

Risk adjustment

• Although six case mix categories appear to be valid for risk adjustment of the measure, NAHC requests that data be made available to enable stakeholders to assess whether such categories are effective in terms of predicting cost.

• Additionally, consistent with our recommendations for the Potentially Preventable Readmission and the Discharge to Community measures, NAHC urges the developer to include in the risk adjustment model the following elements:

Functional status

Caregiver support


Socio-economic status

General Comments:

• The amount of Medicare spending per beneficiary (MSPB) does not have a direct correlation to the quality of care provided by Medicare providers. Although MSPB may provide some useful information regarding resource utilization, it should not be used as the sole resource use measure or to draw assumptions on quality of care on its own. In addition, using MSPB as a quality measure could negatively impact access to home health care for certain high risk patient populations as providers are incented to avoid “high cost” patients in order to achieve favorable scores on the measure.

• Consistent with our comments on the other IMPACT Act measures, NAHC strongly urges testing and validating this measure before it is implemented.

Thank for the opportunity to submit these comments. [Addendum 2/5/2016] This communication is the second of two provided by the National Association for Home Care &

Hospice (NAHC) on the Centers for Medicare & Medicaid Services’ (CMS’) call for input on the Draft Specifications: Medicare Spending Per Beneficiary -- Post Acute Care (MSPBPAC) Resource Use Measures. Our first communication was sent on January 29, 2016. This communication will provide comment exclusively on CMS’ specific request for input on the treatment of episodes with hospice care as part of the measures.

At the outset we believe it is important to note that CMS’ meaning in requesting “input on the treatment of episodes with hospice care as part of the measures” is not entirely clear. Our assumption is that CMS is asking for input on the inclusion of hospice as part of the “Associated Services” during the full “episode” time frame associated with each measure. Our comments reflect that assumption.

As indicated in the Introduction to the Acumen draft specifications report:

“The Improving Post-Acute Care Transformation Act of 2014 (IMPACT Act) authorizes the Secretary to develop ‘resource use measures, including total estimated Medicare spending per beneficiary’ and to require the reporting of standardized assessment data in post-acute care (PAC) settings: skilled nursing facilities (SNFs), home health agencies (HHAs), long-term care hospitals (LTCHs), and inpatient rehabilitation facilities (IRFs). The Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) measures must be implemented according to …statutorily mandated timelines...

“The purpose of the MSPB-PAC measures is to support public reporting of resource use in all four PAC provider settings as well as to provide actionable, transparent information to support PAC providers’ efforts to promote care coordination and improve the efficiency of care provided to their patients.”


As such, it is our sense that the goal of the IMPACT Act and the work that CMS is conducting to implement the Act is to measure the “relative efficiency” of post-acute care interventions, including assessment of the quality and cost of care provided. As noted in the Acumen Draft Specifications, “Associated Services” are suggested for inclusion in the Resource Use measure because these services may be “reflective of and influenced by the services rendered by the PAC facility.”

Hospice is an important element of the health care continuum but it is not post-acute care, and there are many significant distinctions between the reasons for utilization of post-acute care services and referral to and/or election of hospice care. The ultimate goal of post-acute care services in most instances is to either restore a patient to a previous, higher level of functioning or to maximize the potential functioning of the individual so that they are able to resume participation in their lives. Post-acute care requires physician referral for services, and as a rule is initiated with specific or general restorative goals in mind.

While referral to and/or election of hospice services may occur following a hospitalization or post-acute care episode, it does not serve as an alternative site of care or as the “next stage” in the continuum of care for achieving the goals of acute hospitalization or post-acute care. Rather, it represents a change in the focus of care -- an acknowledgement of the presence of a life-limiting illness (with a prognosis of six months or less) and the decision to forego curative care for the terminal illness and related conditions and to pursue palliation and management of these conditions until life ends. This waiver of curative services is, as distinct from every other Medicare benefit, a requirement exclusive to hospice care. Further, while a physician certification is required for hospice care, a patient or family member can pursue hospice coverage without explicit referral to hospice care. In some cases the post-acute care provider may not even know that a patient has subsequently elected hospice care.

We have a number of serious concerns relative to CMS’ inclusion of hospice as part of the “Associated Services” under the post-acute resource utilization measure. We are particularly concerned that inclusion of hospice as an “Associated Service” could deter or significantly delay appropriate referrals to hospice as this may have negative financial implications for the post-acute care or other providers. Additionally, inclusion of hospice as an “Associated Service” would be inappropriate if the terminal/related conditions are not the condition(s) for which referral to post-acute care was made. Therefore, we urge that CMS exclude hospice care from the “Associated Services” category as part of the Resource Use Post-Acute Measure.

We appreciate your thoughtful consideration of our recommendation in this regard.





Name, Credentials, Organization, and Email of Commenter: Heather P. Jones, MPH, CHES,

COS-C, Associate Vice President of Quality Initiatives & State Relations, SC, Association for

Home & Hospice Care of North Carolina/South Carolina Home Care & Hospice Association

[email protected]


TEXT OF COMMENT:

Thank you for the opportunity to submit comments on the development of a Medicare Spending per Beneficiary quality measure for home health providers. The Association for Home & Hospice Care of North Carolina and the South Carolina Home Care & Hospice Association offer the following comments on behalf of our home health agencies.

We have specific concerns that the time period under consideration for home health agencies unlike other post-acute care providers extends well beyond their opportunity for impact. We support the use of a geographic area rather than a national comparison as we believe that to be a more appropriate comparison. We support the inclusion of risk adjustment factors that take into account a Medicare-Medicaid dual eligible population and the impact of socio-economic status. If we can provide any additional information to help support your measure development work, please feel free to contact me.

REPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-20, 24, 30, and 46.



Name, Credentials, Organization, and Email of Commenter: Callene Bentoncoury,

Administrator, Casa de la Luz Hospice, [email protected]

Type of Organization: Hospice

TEXT OF COMMENT:

As we understand it, the MSPB-PAC measures assess a PAC provider’s resource use during an episode. The episode includes not only the time the beneficiary receives services directly from a home health, skilled nursing facility or long term care provider, but also the 30 days following discharge. Included in the resource utilization are hospice care costs for these two time




periods. We strongly believe that including hospice in the measure of resource utilization for PAC has the potential to limit beneficiaries’ access to hospice services.

Hospices around the country are already experiencing a significant increase in very late referrals to hospice care. Here in Arizona, 37% of the patients admitted to our hospice during 2015 died within 7 days of admission. Many PAC providers are already actively treating dying patients to the end of any Medicare payment period in the mistaken assumption that their interventions toward cure are needed. These interventions often deny people the peace and comfort that would better serve them at this time in their lives.

Further incentivizing providers to limit access to hospice care does not make sense. First, people need experts in end of life care to reasonably manage the complexities of death and dying for Medicare beneficiaries and their families. End of life care provided by PAC providers is not the same as that provided by hospice. Second, multiple studies show that use of hospice decreases Medicare costs. A study published in 2015 showed, “Hospice use over 2 weeks was associated with decreased hospital days (1–5 days overall, with greater decreases for longer hospice use) for all beneficiaries.” (Zuckerman, Stearns & Sheingold, .Gerontological Society of America, 2015.) Finally, quality of the end of life experience from the families’ point of view is improved with earlier referral to hospice. According to study published January 2016, “Timely access to hospice care was found to be a major factor in bereaved families' rating the care their loved ones received at the end of life as excellent.” (Wright, Alexi A. MD, et al, JAMA. 2016; 315(3):284-292.)

We highly encourage you to remove hospice care from inclusion in MSPB- PAC measures.

Thank you for your time and consideration.




Name, Credentials, Organization, and Email of Commenter: Cherri Burzynski, MSN, RN,

NE-BC, President, National Association of Long Term Hospitals, [email protected]


TEXT OF COMMENT:

The National Association of Long Term Hospitals (NALTH) is pleased to submit comments on the Medicare Spending per Beneficiary measure for the Post-Acute Care (PAC) Long-Term Care Hospital Quality Reporting Program (Required under the IMPACT Act). NALTH is the only hospital trade association in the nation that is devoted exclusively to the needs of patients who require services provided by long term care hospitals (LTCHs). NALTH is committed to



research, education and public policy development that further the interests of the very ill and often debilitated patient populations who receive services in LTCHs throughout the nation. NALTH’s membership is composed of the nation’s leading LTCHs, including free-standing, hospital-within-hospital, for-profit, and non-profit LTCHs. On behalf of our member hospitals, we wish to express our gratitude for the opportunity to share our comments on the draft specifications for the MSPB-PAC measures.

While we support the goal of evaluating providers’ resource use relative to the resource use of the national median provider, we have identified several concerns regarding the ability of the measure to accurately reflect resource use differences across LTCHs. We discuss these concerns below.

General Comments

Cross-Setting Comparisons. We appreciate that “each MSPB-PAC measure only compares providers within a given PAC setting” and that “different types of PAC providers are not compared to one another.” We echo the concerns of several of the panelists during the October 29 and 30, 2015 Technical Expert Panel on the MSPB-PAC measures that current data and methods do not permit comparisons across different types of PAC providers on spending over an episode of care. We agree that “given the lack of standardized assessment data, as well as inherent differences in payment systems and patients populations across PAC settings”, CMS should “undertake considerable research and gather substantial stakeholder input if the measures were to be adapted for this purpose in the future.”

LTCHs differ from other PAC settings in important ways. First, LTCHs, along with inpatient rehabilitation facilities, must meet the requirements of an acute care hospital. Second, LTCHs must have an average length of stay of more than 25 days. As a result, LTCHs treat higher severity cases than other post-acute care settings. In 2006, approximately 37% of LTCH cases grouped to the highest APR-DRG severity score, while this percent ranged from 4% to 7% for other post-acute care (PAC) providers. (Koenig et al. The Effects of Long-term Care Hospitals on Outcomes, Utilization and Payments for Medicare Beneficiaries. November 7, 2013. Final Report prepared for the National Association of Long Term Hospitals) Patients treated in LTCHs often possess multiple comorbidities and require specialized care. For example, 28.0% of LTCH patients with digestive system problems had at least three major complications or comorbidities compared to 2.2% of patients with digestive system problems in other PAC settings. (Lane Koenig, Berna Demiralp, Josh Saavoss, and Qian Zhang, “The Role of Long-term Acute Care Hospitals in Treating the Critically Ill and Medically Complex: An Analysis of Nonventilator Patients,” Medical Care 53(7) (July 2015): 585) As another example, LTCHs frequently treat patients on prolonged mechanical ventilator with the purpose of weaning the patient. Few other PAC providers see such patients. We do not believe that claims data alone are sufficient to appropriately adjust or control for severity differences between LTCHs and other providers. Thus, we strongly support separate measures for each of the PAC settings. In addition, CMS must make clear when the data become available for public display that it is inappropriate to compare MSPB-PAC measure values across settings.

MSPB is Not a Quality Measure: On page 4 of the measure specification (Introduction), the document states “[g]iven the design of the MSPB-PAC measures, post-treatment costs may serve as an indicator of the quality of care provided during PAC care, in that higher quality PAC


treatment may yield lower post-treatment costs.” We disagree with this statement. Costs by themselves are not an indication of quality and should not be interpreted as such. Doing so can result in negative consequences for Medicare beneficiaries. Costs must be interpreted in combination with clinical outcomes. Since the MSPB does not include clinical outcomes, it cannot by itself reflect quality.

In a number of places, the document states that the MSPB-PAC measures reflect provider efficiency. Again, we disagree and ask that Acumen and CMS refer to MSPB-PAC measures as resource use measures and no more. True efficiency measures, would compare costs given a specific level of quality. However, the MSPB-PAC measures do not include or reflect quality.

Issues regarding LTCH interrupted stays

The specifications do not discuss how LTCH interrupted stays will be treated. While services received for interrupted stays for three days or less are the responsibility of the LTCH, services received during interrupted stays of more than three days are included in the claim for the short term acute care hospital (STCH), IRF, or SNF. If the LTCH stay is a site-neutral episode, the interrupted stay is only counted during the treatment period and the specifications do not clarify if the STCH, IRF, or SNF claims will be used to capture treatment services during the interrupted stay. We suggest that the measure development team clarify how these interrupted stays will be treated.

Concerns with including short LTCH stays

We are concerned that the MSPB-PAC measure will hold LTCHs accountable for very short LTCH stays. For example, consider a patient that is discharged from a STCH to an LTCH, stays in the LTCH for a couple days, has an acute exacerbation that requires acute/intensive care, and then is readmitted to the STCH. The LTCH stay will trigger a new episode with the spending during the LTCH stay included as treatment services and the spending during the readmission to the STCH included as associated services. However, since LTCHs are defined, in part, based on having an average length of stay of more than 25 days, these short stays may represent episodes where the STCH discharged the patient too early. The LTCH should not be accountable for the patient in these instances. We ask the measure development team to consider establishing a minimum number of days in an LTCH or other setting before an episode counts toward the MSPB-PAC measure.

Issues Related to Risk Adjustment

In comparing between LTCH facilities, we are concerned that the current risk adjustment variables will not adequately capture patient differences. Without sufficient risk adjustment, differences in spending may be due to differences in patients’ clinical characteristics and may not be attributable to differences in resource use across providers.

We recommend adjusting for additional factors that play a significant role in determining resource use in the LTCH setting, such as prolonged mechanical ventilation, multiple organ failure (beyond those included in the variable interaction terms), and the number of days in an intensive care unit. We also recommend adjustment based on socioeconomic factors.


Even though a separate risk adjustment model will be estimated for episodes with the most recent institutional claim being a prior PAC claim, we recommend adding the type of prior PAC setting as a risk adjustor. Patients with a prior HHA claim and patients with a prior LTCH claim may differ greatly in terms of their acuity.

While we support estimating separate risk adjustment models for the six clinical case-mix categories described on pg. 21 and a separate model for LTCH site-neutral cases, we are concerned that the sample sizes in each of the clinical case-mix categories will lead to imprecise risk adjustment. It is important that the measure development team present statistical analyses showing the effects of using these clinical case-mix categories on the precision of the predicted episode payments from the risk adjustment model as well as estimates of the reliability of the estimates.

While Medicare claims data are more readily available than other data sources, they may not capture finer distinctions across patients that may affect the patients’ outcomes and facility to which they are discharged. Therefore, a process to include assessment data in the MSPBPAC measure calculations, once available, needs to be established and followed.

Concerns regarding the public comment process

While we appreciate that the comment period was extended by two days, the period provided for public comment is insufficient to provide complete comments regarding these complex measures that may have profound implications for NALTH members. We encourage CMS and the measure development team to provide at least 30 days for the comment periods in the future.

In addition, the specifications do not provide enough detail for a convincing argument on some of the choices made by the measure development team. For example, collapsing readmissions for the same patient and providers within 7 or fewer days as part of the same treatment period may be reasonable, but the measurement team provides sparse support for this decision in their appendix D. In addition to listing the types of empirical analysis conducted, we urge the measurement team to report the results from these analyses so the public can make their own conclusions regarding the analyses and analytic choices made by CMS.

If you have any questions about these comments, please contact [contact details]

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 3, 4, 9, 17, 19, 21, 30, 34, 39, 41, and 42.



Name, Credentials, Organization, and Email of Commenter: Theresa L. Lee, JD, MPH,

Executive Director, Alliance for Home Health Quality and Innovation [email protected]




TEXT OF COMMENT:

The Alliance for Home Health Quality and Innovation (the “Alliance”) appreciates the opportunity to comment on the draft specifications for the Medicare Spending Per Beneficiary—Post-Acute Care (MSPB-PAC) resource use measures.

By way of background, the Alliance is a non-profit 501(c)(3) organization with the mission to lead and support research and education on the value of home health care to patients and the U.S. health care system. Working with researchers, key experts and thought leaders, and providers across the spectrum of care, we strive to foster solutions that will improve health care in America. The Alliance is a membership-based organization comprised of not-for-profit and proprietary home health care providers and other organizations dedicated to improving patient care and the nation’s healthcare system. For more information about our organization, please visit: http://ahhqi.org/.

The Alliance has the following comments regarding the proposed measures.

First, the Alliance appreciates the importance of developing measures to better understand Medicare cost and resource use associated with post-acute care, but is concerned that when viewed in isolation, cost information alone can be a confusing measure because it does not necessarily correlate with quality of care. The Alliance believes that spending alone is not an indicator of quality, nor is it an indicator of efficiency. The measure will be most useful when paired with quality outcome measures. If outcome measures are not linked to this cost measure, there may be an incentive for providers not to refer patients for reasonable and necessary services, including post-acute care services that can be used to reduce rehospitalization rates and improve patient experience. As various stakeholders make use of MSPB-PAC measures and apply them in various ways, it will be critical to ensure that adequate quality outcome measures are coupled with measures of the cost of care to discourage underuse.

Moreover, there is a need for the development of measures of patient access to care. Reforms that are aimed at improving efficient, cost-effective delivery of care are needed, as are measures that will help to encourage efficiency. However, patient access should not be compromised as a means to lower cost. Unfortunately, this measure alone cannot be used to assess whether patients have access to quality care. The Alliance is not aware of any measures at present that would address access to care and would support development of such measures.

Second, the Alliance is concerned about the episode length for the MSPB-home health measure as it relates to standard HHA and HHA-LUPA episodes. Whereas for SNF, LTCH, IRF and HHA-PEP episodes, the treatment period ends at discharge, for standard HHA and HHA-LUPA episodes the treatment period ends after 60 days. For standard HHA episodes, there are cases when a beneficiary is discharged prior to the end of the 60 day episode. For example, a beneficiary may be discharged at the 25th day for a variety of reasons (e.g., the patient may no longer be homebound, or the patient’s goals have been met). In such cases when the standard HHA episode ends before the 60th day, the treatment period should end at discharge instead of running a full 60 days. Similarly, for HHA-LUPA episodes, the treatment period should end at discharge because in many cases patients may be discharged before the 60th day.

By making this change, the home health treatment periods will be better aligned with the way the other PAC provider treatment periods are constructed in terms of approach and methodology.

http://ahhqi.org


Third, the Alliance is concerned that for the MSPB-home health measure, the home health costs will be double counted because of the way the episode length is constructed. In cases where there is a home health episode followed by recertification and a second home health episode, the second home health episode of care would be counted twice: once as part of a treatment period associated with the second home health episode and another as part of the first home health episode’s associated services period. The Alliance urges the measure developer to ensure that home health episode costs will not be counted twice. This is an issue that may be unique to home health care and the Alliance recommends that the episode length be tailored to prevent overlap and over-counting.

Fourth, the Alliance supports the overall approach to focusing on MSPB within each PAC setting. This will enable providers in each PAC setting the opportunity to test the measures and assess whether refinements to the measures are needed. After an appropriate period of testing, it may make sense over time to consider development of a measure that will enable comparisons across post-acute care settings. If the policy interest in post-acute care payment reform results in a unified post-acute care payment system across settings, a cross-setting MSPB measure might be considered in the future to align with such a system.

Fifth, the Alliance is concerned about use of the national median as the benchmark for the MSPB-PAC measures. At present, there is considerable geographic variation in health care that is often outside the control of any given provider. For example, in some areas, there is a high penetration of Medicare Advantage plans and the patients in traditional Medicare tend to be older and more sick or frail than MA plan beneficiaries; in these areas, resource use per beneficiary in traditional Medicare may be higher than expected. Notwithstanding, the Alliance recognizes that it is a reasonable policy goal to achieve greater consistency in practice where variation is inefficient.

The Alliance recommends using regional medians as benchmarks. It is worthy of note that in the CMS comprehensive care for joint replacement model, CMS is using a phased in approach for its cost benchmarks that are blended provider-specific costs and regional medians for the first four years, and shifting to straight regional medians in year five. A similar approach could be used for the MSPB-PAC measures. Alternatively if a national median must be attained CMS should use a similar blended approach using regional and national medians for the transition period and ultimately shifting to straight national median.

Sixth, it is interesting to note the varying approaches to risk adjustment that have been used by the measure developers of the different IMPACT Act domain measures. The Alliance supports recognition of prior PAC setting as a factor relating to risk adjustment and appreciates this approach. Regarding the six different groupings for risk adjustment specified in the measure specifications, the Alliance respectfully requests that data be made available to enable stakeholders to assess whether such categories are effective in terms of predicting clinical similarity and cost. The Alliance also recommends the use of approaches to risk adjustment that make use of data (including assessment data) collected on functional status and behavioral and mental health. It will also be critical to pursue and invest in ways to risk adjust for socioeconomic status.

Finally, the Alliance urges testing, validating and self-reporting of this measure before it is finalized. Testing and validation should be no less than six months with an opportunity to modify


the measure prior to finalizing it. A similar approach was used for many of the OASIS-based measures that CMS uses for home health agencies. The information on performance against this measure should not be made public if there are any concerns about validity and accuracy of the measure.

The Alliance appreciates the opportunity to comment on the draft MSPB-PAC measure specifications.


Regarding your comment about the need for access to care measures, access to quality care is a top priority for CMS. CMS will take the suggestion for the development of measures assessing access to care under consideration.



Name, Credentials, Organization, and Email of Commenter: Scott Laker, MD, Chair, Health

Policy & Legislation Committee, American Academy of Physical Medicine and Rehabilitation

[email protected]


TEXT OF COMMENT:

The American Academy of Physical Medicine and Rehabilitation (“AAPM&R” or “the Academy”) appreciates the opportunity to submit comments to Acumen, LLC on its document entitled, “Draft Specifications for the Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) Resource Use Measures.”

AAPM&R is the national medical society representing more than 8,000 physiatrists, physicians who are specialists in the field of physical medicine and rehabilitation. Physiatrists treat adults and children with acute and chronic pain, persons who have experienced catastrophic events resulting in paraplegia, quadriplegia, traumatic brain injury, spinal cord injury, limb amputations, rheumatologic conditions, musculoskeletal injuries, and individuals with neurologic disorders or any other disease process that results in impairment and/or disability.

Physiatrists coordinate, supervise and provide medical rehabilitation services in a wide variety of settings including all of the post-acute care settings impacted by these draft specifications. While physiatrists have had a close affiliation with Inpatient Rehabilitation Hospitals and Units for decades, physical medicine and rehabilitation (PM&R) physicians are increasingly present across the post-acute care continuum. As such, physiatrists are not aligned with any one PAC setting and, as a result, can act as an impartial medical decision-maker to help direct patients to



the most appropriate setting and intensity of rehabilitative care to meet the individual medical and functional needs of patients.

There may seem to be some incongruity for a physician society to comment on a set of facility-based PAC quality measures, but the fact is that physiatrists and members of the rehabilitation provider team are the professionals serving patients in these settings. While the measures themselves reflect the facilities’ performance and may well lead to financial consequences for these facilities based on performance under these measures, it is physicians and other rehabilitation professionals who ultimately determine the care decisions made in these PAC settings.

With this in mind, we offer a series of specific comments for consideration by Acumen, LLC.

• AAMP&R believes that the comment period is simply too short to offer meaningful and in-depth comments on these complex measures. The call for public comment opened on January 13, 2016 and closes on January 29, 2016. We object to such a quick comment period. Due to these time constraints, our comments will be brief.

• As a matter of principle, using Medicare Spending Per Beneficiary (MSPB) measures for PAC can very easily lead to discrimination of providers that have patient populations with greater than average medical needs. While providing much needed care for these patients, such providers will appear to be inefficient under an MSPB payment model.

• These same forces may prompt some PAC providers to cherry-pick patients and divert the most costly or complex patients away from admission to their facilities, unless strong patient safeguards are in place.

• That being said, we appreciate that the MSPB measures are proposed in separate care setting episodes (home health agency (HHA), skilled nursing facility (SNF), long-term care hospital (LTCH), and inpatient rehabilitation facility (IRF)) rather than across post-acute care. Such an approach better takes into account different patient populations and care needs and makes more relevant comparisons.

• The exclusions from an MSPB model are critical, and at a minimum, Durable Medical Equipment, Prosthetics, Orthotics and Supplies (DMEPOS) and End-Stage Renal Disease (ESRD) patients should be excluded.

o Patients in need of DMEPOS should be excluded because custom orthotics, all prosthetics, and complex wheelchairs are relatively expensive items. Including them in an MSPB model will encourage providers to delay care so as not to provide them during their stay. This will pass costs on to post-discharge settings, and delay, if not deny altogether, clinically appropriate care.

o ESRD patients should be excluded as these patients are very few in number, but are by nature high utilizers of care. The expense of treating patients with ESRD may prompt some providers to avoid ESRD patients, or prejudice those providers who do admit them and incur significant costs of care.


AAPM&R appreciates the opportunity to offer these comments and looks forward to working with CMS and its contractors as implementation of the IMPACT Act continue.

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-4, 8, 9, 28, and 29.



Name, Credentials, Organization, and Email of Commenter: Barbara A. McCann, Chief

Industry Officer, Interim HealthCare [email protected]

Type of Organization: HHA

TEXT OF COMMENT:

We appreciate the time and analysis required to think through such a complex measure. As Medicare HHA certified providers in over 100+ locations across the country, we would like to offer the following comments for your consideration.

Support and Caution:

• We support the recommended comparison of the LUPA episodes to other LUPA episodes, as well as comparing PEP episodes to other PEP episodes. These elements of the home health prospective payment have an important role in delivering care, however, such an analysis will, we hope, inform the industry and others as to the role they actually play within this payment model.

o Our caution relates to the industry ignorance about the data and the opportunity that exists for the industry to learn from its inclusion. It has been collected for over a decade and never really offered to the industry. The measures addition of more information than a comparative ratio we believe would enhance the publication of the comparative rates and support the stated goal of improving coordination and efficiency. We recommend that such information include the profile of these patients; operational information such as did the PEP occur due to a change in an agency-not uncommon for a community based HHA patient to be admitted and then discharged to the hospital-based agency in an era of intense competition, etc.

o A second caution relates to LUPAs and HHAs, namely that although there may be an ongoing medical need, the patient must be discharged when they no longer are homebound. How will the random interpretation of this element of benefit eligibility be taken into consideration in the analysis? Or is there a way to identify it as a factor? It could be informative in the debate about this requirement of the benefit.

• We support the definition of the episode window, and both elements of the treatment period and the associated services.

o Our caution is that the simple publication of the ratios will not alone achieve the purpose of the measure. A great deal of education will be needed.



• We support the definition of the exclusions, with one caution. Unlike facilities where there is assumed responsibility for 24/7 care, that is not the case with HHAs. Therefore in the 30 days of associated services or even during the treatment period, enrollees may see multiple other doctors, none of which are associated with the doctor ordering HHA and overseeing the care plan. We often do not know about such actions. Our caution is how to fairly analyze the Medicare costs when the patient receives care in the community and has open choice to access and use Medicare FFS services which is not the case in other PAC settings-namely they can ‘control’ the costs incurred.

o Caution, with the stated purpose of the measure and extended periods for Medicare cost inclusion during the “associated services” period, can you please consider that HHAs do not even know what other costs are being incurred unless related to the physician who is providing oversight. We cannot improve coordination or communication if there is no provision of information to take action -perhaps at least annually the names of the providers and related other services during the treatment period and associated services period could be offered on CASPER or another secure portal to initiate the goal of this measure.

• The consideration of each final claim for HHAs we applaud. We remain concerned as do others in the industry about the group of patients that continue episode after continuous episode.

o Our concern relates to separating fraud from ongoing M&E episodes for those in end stage illness and who have not selected hospice care.

o Our second concern relates to unique elements of the home health benefit and how such will be considered in the analysis. One such patient population are those for whom we provide catheter care or changes monthly. They may be evident most often in LUPAs as 4 or less visits are provided. They are usually with the HHA for 4 or more continuous episodes. In our own analyses we can only identify by looking for the primary or secondary diagnosis code that relates to neurogenic bladder or managing urinary devices. The benefit was created to keep these individuals at home with only the care needed-how will this unique patient population be accounted for in the analysis?

• When normalizing of the HHA claims is conducted, will ‘outlier’ payments as defined in HHA PPS be included - namely high expenditure patients. We would support such inclusion as it is indicative of several issues in managing patient care; this relates specifically to Step 6.

• We are confused as to why hospice costs are included in the associated services for HHAs. The patient accepted and was transferred to another setting. We do not understand why these costs would be in ‘associated services’ especially as under the new hospice payment system the cost per day would be higher as it would be early in the hospice episode. Also, hospice payment is /day – very different than HHA PPS and includes medications, inpatient care, etc. We remain respectfully confused how these can be incorporated in the post 30 day period or during the 60 days. Could hospice costs at least be labeled as such? Please note that one of our goals is advance care planning which can include supporting end stage patients to transferring to hospice earlier, but now we would be penalized for high costs during the 60 days or 30 days thereafter?

• With regard to risk adjustment and the categories proposed, we support those noted, but would recommend an additional category – namely end stage or palliative care which often does not involve ICU and is very different than general med/surg. As patients begin the


dying process and do not accept hospice care, there is often an intense sequence of ER visits, admissions, discharges, followed by death. These are end stage patients, how is that managed in the risk adjustment? Is it solely managed by exclusion if death occurs within a certain timeframe of a PAC setting admission? We recommend that such patients be included in a category for the purpose of risk adjustment as we believe these patients and their related costs serve an important purpose of the measure to improve coordination across sites and better manage costs and patients.

Other General Comments and Concerns:

• The absence of a “Definitions” section made the measure difficult to understand. • In particular, we could not find the definition of “PB”. • The ‘long term care indicator’ is poorly defined. • Is the claim limited to its status at the time the measurement occurs, in other words it will

consider denials, recoupments, etc.? • Does normalizing remove the rural add-on for HHAs? • P. 20 – we cannot understand the ‘lookback’ period as it relates to death. Can that please be

clarified?

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-1, 5, 11, 13, 14, 15, 24, and 27.

Thank you for your support of the HHA LUPA to HHA LUPA, and HHA PEP to HHA PEP comparison. Regarding your comment on MSPB-PAC HHA LUPA episodes, these are compared only to other LUPA episodes. This ensures that the beneficiaries (e.g., requiring catheter care) are compared to other similar patients with MSPB-PAC HHA LUPA episodes. The definition and requirements of the HH PPS apply uniformly to the scenario that you suggest where a patient is no longer homebound and is therefore discharged, with ongoing medical need.

Regarding your comment on outlier payments, they are included as a part of the standardized allowed amount for HHA claims.

Regarding your comment on fraud, the MSPB-PAC measures are resource use measures and are not designed to identify health care fraud.

Regarding your comment on end stage patients, as noted, we exclude episodes in which beneficiaries die during the episode window. We acknowledge the importance of measuring resource use in episodes involving beneficiaries who have the sequence of care described in the comment. Our concern is the feasibility and reliability of using claims data to identify end-stage patients who refuse hospice treatment, for the purposes of risk adjustment. (Please see Comment Summary ID-45 for a discussion of claims reflecting acuity.) An alternative approach would be to include episodes ending in death in the measure and risk adjust using an indicator variable for any episode ending in death. This poses other challenges, such as the heterogeneity in timing of death -- and hence in completeness of episodes and resource use -- among these episodes. Recognizing such challenges, we have presently chosen to exclude episodes ending in death, as is the case with the Hospital MSPB measure.


Regarding your comments under ‘general comments and concerns’, PB is defined on page 8 of the draft measure specifications document: "The treatment period is the time during which the patient receives care services from the provider for whom the measure is being calculated (the “attributed PAC provider”), and includes claims for the PAC provider as well as all Physician/Supplier (PB) and durable medical equipment, prosthetics, orthotics, and supplies (DMEPOS) claims, excepting services that are determined to be clinically unrelated to PAC treatment." However, we recognize that terminology is important and propose to include a glossary for terms used throughout the measure specification document. Please see Summary Comment ID-11 for discussion of glossary and the definition of the long-term care indicator.

We standardize payment to account for payment differences attributable to geographic differences as well as incentive payment adjustments and other add-ons. This is the same payment standardization used for the NQF-endorsed hospital MSPB measure. Specifically, 50 percent of the rural home health add-on is applied to all home health episodes, rural and urban. This ensures that episodes for rural home health agencies do not appear more expensive due to the rural add-on and that rural and urban home health episodes are comparable.

The MSPB-PAC measures exclude all episodes for beneficiaries whose deaths fall within the episode window or lookback period prior to the episode start. Please see Summary Comment ID-43 for a discussion of the lookback period.



Name, Credentials, Organization, and Email of Commenter: Cathy Day, RN, C-OB, MSN,

CNML, CJCP, Individual [email protected]


TEXT OF COMMENT:

The MSPB measure puts the general public at risk and should be re-considered. It is a patient’s right to have the course of treatment in which they want (and qualify for) but in the same token the hospital will be penalized for offering those additional services to the patient because of the costs associated with it. I think the measure is good for keeping costs under control as long as it is adjusted based on geographic location as big urban systems can negotiate better pricing for supplies and equipment than smaller, independent rural organizations can.


Regarding your comment about risks, the MSPB-PAC measures are defined in such a way as to incentivize providers to deliver high-quality care, as well as appropriate discharge planning and post-discharge care coordination so that beneficiaries would likely experience fewer costly



adverse events. Please see Summary Comment ID-24 for a discussion of the associated services period.



Name, Credentials, Organization, and Email of Commenter: Kira M. Carter-Robertson,

MHA, FACHE, VP, Post-Acute Services, Sparrow Health System [email protected]


TEXT OF COMMENT:

On behalf of Sparrow Specialty Hospital (SSH), I am submitting comments on the Medicare spending per Beneficiary measure for the Post-Acute Care (PAC) Long-Term Care Hospital Quality Reporting Program (Required under the IMPACT Act).

Sparrow Specialty Hospital is a 30 bed long-term care hospital (LTCH) located in Lansing, Michigan which provides specialized programs of care to chronically critically ill and medically complex patients. SSH is an affiliate of Sparrow Health System, Lansing, Michigan. We serve a significant percentage of Medicare patients residing in the Grater Lansing Area. We are appreciate of the opportunity to provide comments on this proposed rule.

While we support the measure’s goal of evaluating providers’ resource use relative to the resource use of the national median provider, we have identified several concerns regarding the ability of the measure to accurately reflect resource use differences across LTCHs. We discuss these concerns below:

1) MSPB is Not a Quality Measure: On page 4 of the measure specification (Introduction), the document states “given the design of the MSPB-PAC measures’ post-treatment costs may serve as an indicator of the quality of care provided during PAC care, in that higher quality PAC treatment may yield lower post-treatment costs.” We disagree with this statement. Costs by themselves are not an indication of quality and should not be interpreted as such. Doing so can result in negative consequences for Medicare beneficiaries. Costs must be interpreted in combination with clinical outcomes. Since the MSPB does not include clinical outcomes, it cannot by itself reflect quality. In a number of places, the document states that the MSPB-PAC measures reflect provider efficiency. Again, we disagree and ask that Acumen and CMS refer to MSPB-PAC measures as resource use measures and no more. True efficiency measures, would compare costs given a specific level of quality. However, the MSPB-PAC measures do not include or reflect quality.

2) Cross-Setting Comparisons. We appreciate that “each MSPB-PAC measure only compares providers within a given PAC setting” and that “different types of PAC providers are not compared to one another.” We echo the concerns of several of the panelists during the October 29 and 30, 2015 Technical Expert Panel on the MSPB-PAC measures that current data and methods do not permit comparisons across different types



of PAC providers on spending over an episode of care. We agree that “given the lack of standardized assessment data, as well as inherent differences in payment systems and patients populations across PAC settings”, CMS should “undertake considerable research and gather substantial stakeholder input if the measures were to be adapted for this purpose in the future.” LTCHs differ from other PAC settings in important ways. First, LTCHs, along with inpatient rehabilitation facilities, must meet the requirements of an acute care hospital. Second, LTCHs must have an average length of stay of more than 25 days. Thirdly, LTCHs frequently treat patients on prolonged mechanical ventilator with the purpose of weaning the patient. Few other PAC providers see such patients. We do not believe that claims data alone are sufficient to appropriately adjust or control for severity differences between LTCHs and other providers. Thus, we strongly recommend separate measures for each of the PAC settings. In addition, CMS must make clear when the data become available for public display that it is inappropriate to compare MSPB-PAC measure values across settings.

Additional Concerns:

• LTCH interrupted stays – The specifications do not discuss how LTCH interrupted stays will be treated. While services received for interrupted stays for three days or less are the responsibility of the LTCH, services received during interrupted stays of more than three days are included in the claim for the short-term acute care hospital (STCH), IRF, or SNF. If the LTCH stay is a site-neutral episode, the interrupted stay is only counted during the treatment period and the specifications do not clarify if the STCH, IRF, or SNF claims will be used to capture treatment services during the interrupted stay. We recommend that the measure development team clarify how these interrupted stays will be treated.

• Short Stay LTACH Stays – We are concerned that the MSPB-PAC measure will hold LTCHs accountable for very short LTCH stays. For example, consider a patient that is discharged from a STCH to an LTCH, stays in the LTCH for a couple days, has an acute exacerbation that requires acute/intensive care, and then is readmitted to the STCH. The LTCH stay will trigger a new episode with the spending during the LTCH stay included as treatment services and the spending during the readmission to the STCH included as associated services. However, since LTCH are defined, in part, based on having an average length of stay of more than 25 days, these short stays may represent episodes where the STCH discharged the patient too early. The LTCH should not be accountable for the patient in these instances. We recommend the measure development team consider establishing a minimum number of days in an LTCH or other setting before an episode counts toward the MSPB-PAC measure.

• Issues Related to Risk Adjustment – In comparing between LTCH facilities, we are concerned that the current risk adjustment variables will not adequately capture patient differences. Without sufficient risk adjustment, differences in spending may be due to differences in patients’ clinical characteristics and may not be attributable to differences in resource use across providers. We recommend adjusting for additional factors that play a significant role in determining resource use in the LTCH setting, such as prolonged mechanical ventilation, multiple organ failure (beyond those included in the variable interaction terms), and the number of day s in an intensive care unit. We also adjustment based on socioeconomic factors. In addition, even though a separate risk adjustment


model will be estimated for episodes with the most recent institutional claim being a prior PCA claim, we recommend adding the type of prior PAC setting as a risk adjustor. Patients with a prior HHA claim and patients with a prior LTCH claim may differ greatly in terms of their acuity. While we support estimating separate risk adjustment models for the six clinical case-mix categories describe on pg. 21 and a separate model for LTCH site-neutral cases, we are concerned that the sample sizes in each of the clinical case-mix categories will lead to imprecise risk adjustment. We recommend that the measure development team present statistical analyses showing the effect of using these clinical case-mix categories on the precision of the predicted episode payments from the risk adjustment model as well as estimates of the reliability of the estimates. We recognize that while Medicare claims data are more readily available than other data sources, they may not capture finer distinctions across patients that may affect the patients’ outcomes and facility to which they are discharged. Therefore, a process to include assessment data in the MSPB-PAC measure calculations, once available, needs to be established and followed.

Thank you once again for the opportunity to provide comments on the Medicare Spending per Beneficiary measure for the Post-Acute Care (PAC) Long-Term Care Hospital Quality Reporting Program (Required under the IMPACT Act).

RESPONSE: Thank you for your comment. Please see our response in Section 2, at Summary Comment IDs-2, 13, 19, 21, 34, 39, and 42.



Name, Credentials, Organization, and Email of Commenter: R. Claiborne Richards, Jr. Chief

Executive Officer, naviHealth [email protected]


TEXT OF COMMENT:

We want to thank CMS and its measure contractor for the opportunity to comment on the Draft Specifications for the Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC). Generally, we support CMS’ efforts to shift to value-based payments and person-centered care, and we commend CMS’ ongoing efforts to implement a resource use measure for Post-Acute Care (PAC) providers. Incentivizing high quality, efficient care encourages better care, more coordinated care for patients across providers.

naviHealth has years of experience partnering with providers to manager care more efficiently from the acute setting to post-acute settings. While there has been much progress, we acknowledge that there is still training and learning that is necessary before care is seamlessly



coordinated in all settings. We support programs that provide actionable information and raise awareness about inefficient care.

While we agree providers should be measured based on their performance, measures must be defined in a way that does not penalize providers for conditions that are out of their control. It is crucial that the MSPB-PAC appropriately risk adjusts patients to ensure access and appropriate care for the most complex patients. We support ongoing work to improve risk adjustment and the inclusion and exclusion criteria to ensure appropriate attribution of spending as it relates to a PAC episode of care.




Name, Credentials, Organization, and Email of Commenter: Kimberly M. Gimmarro,

Executive Assistant/Quality Specialist, Botsford Commons Senior Community

[email protected]


TEXT OF COMMENT:

Inclusive of a home health component in an episode of care, how will these measures address clients who choose not to allow the home health service to provide post-acute discharge care? There needs to be recognition that, aside from the medical need perspective, an element of trust in allowing total strangers into the home must be present to provide home health. For those patients in particular who are lacking a personal caregiver, family member, friend, or neighbor to be present with them to reassure that the home health worker seeking entry to the patient's home is in fact just that - the issue becomes how does the measure respect patient choice in this regard?

Might an exclusion be noted for home health refused at time of home health visit to the home? Might the measure encourage x number of attempts by the home health service to fulfill the physician-requested first visit?

Providers, in our case a SNF provider, schedule home health with discharge, but scheduling it is one thing - there is an element of uncertainty with the homeowner controlling access to the home in the post-discharge scenario.






Name, Credentials, Organization, and Email of Commenter: Mary Ann Starbuck, Chief

Executive Officer, CareFirst, [email protected]


TEXT OF COMMENT:

Thank you for the opportunity to submit comments on the MSPB – PAC Resource Use Measures. We applaud the effort to develop measures to support public reporting of resource use in all four PAC provider settings as well as to provide actionable, transparent information to support PAC providers’ efforts to promote care coordination and improve the efficiency of care provided to their patients.

Our primary concern is that in the set of proposed measures, for each PAC provider hospice is considered an “associated services” provider. We are very concerned that the inclusion of hospice as an associated services provider will result in patients who are eligible for hospice and would benefit from hospice care being denied timely and appropriate access to hospice services. This concern stems from a realistic view of care from each of the four PAC provider settings. Specifically, we believe:

Hospice meets the standards to be excluded as a clinically unrelated service.

• 3.1.5 Step 5: Excluding Clinically Unrelated Services Certain services are excluded from the treatment and associated services periods because they are clinically unrelated to PAC care and/or because PAC providers may have limited influence over certain Medicare services delivered by other providers during the episode window. Inclusion of services that cannot be reasonably managed by the PAC provider could create incentives for providers to avoid treating patients with certain conditions or complex care needs (e.g., patients requiring chemotherapy or dialysis) that cannot be fully accounted for in risk adjustment models. Further, including such services in the MSPB-PAC measures would limit the extent to which MSPB-PAC measures reflect meaningful and actionable variation in PAC providers’ efficiency.

• Hospice is a unique provision of care with eligibility being a life expectancy of six months or less if the disease runs its normal course. Beyond the referral for service, the PAC providers have no influence over hospice service. Nothing the PAC does would impact the need or eligibility for hospice service. It would be extremely difficult to impossible to fully account for terminal illness and hospice care in risk adjustment models.

• Hospice patients have complex care needs specifically targeted to end of life. Hospice is uniquely designed to efficiently and effectively address these needs:

o expertise in end of life care symptom control



o an interdisciplinary team approach recognizing that end of life symptoms often stem from spiritual and psycho-social as well as grief issues

o awareness of particular pharmacology issues are end of life o provision of care to both the patient and the family or support system of the

patient recognizing that the terminally ill patient is greatly impacted by the family or support system

o A CMS report itself evidences the complexity of the Hospice Medicare benefit in comparison with Home Health Medicare benefit. In the report, Medicare Hospice Payment Reform: Analysis of How the Medicare Hospice Benefit is Used (HHSM-502005-00018I December 3, 2015 Section 10 pp71-75), the disparity in both visits and minutes of service is outlined without even including the coverage of medications, DME, supplies, transportation and more in the Hospice Medicare Benefit while it is separate in the Home Health Medicare Benefit.

o Inclusion of hospice as an associated service will actually impede the measurement of PAC provider efficiency. Hospice is consistently proven to be a cost efficient service for terminally Medicare patients particularly with a 30 day look-back period, even by CMS’s own consultants (Medicare Hospice Payment Reform: Analysis of How the Medicare Hospice Benefit is Used HHSM-502005-00018I December 3, 2015 pg. 27). Yet by including hospice as an associated service, the PAC provider will be penalized for a referral to hospice by the hospice costs that are associated.

• The MSPB-PAC measure creates a disincentive for referral to hospice by PAC providers because these providers are likely to be concerned that the cost of hospice will have a negative impact on their performance on the measure. This disincentive will result in denial of access to hospice services through a variety of pathways :

o PAC providers may not refer patients who are terminally ill to hospice and will either continue to care for them without the needed end of life expertise or simply discharge them. Hospice provides quality, compassionate care for people facing a life-limiting illness. The hospice interdisciplinary team is uniquely equipped to deliver expert medical care, pain management, and emotional and spiritual support expressly tailored to the patient's needs and wishes. Other PAC settings are not able to offer the same level of expertise and holistic comprehensive end-of life care that hospice provides. PAC provider failure to refer patients to hospice because of concerns related to the MSPB-PAC measure will result in denial of access to quality end-of-life care for many patients.

o PAC providers may wait to refer patients to hospice until death is imminent. While hospice providers are able to provide quality care for patients admitted to hospice services close to death, a very short length of service often means that patients and their families are not able to receive all of the available benefits of hospice care. The Medicare hospice was designed to provide comprehensive, interdisciplinary care for beneficiaries in the last six months of life. When hospice care is initiated close to death, patients and families are often in crisis. And while these patients and families still greatly benefit from hospice care, they could benefit even more from a longer length of service – particularly from


psychosocial and spiritual care which utilizes therapeutic relationships and volunteer services that are developed over time.

o PAC providers may also delay referral to hospice until after the 30 day associated services period has elapsed. A 30 day delay would deny timely access to hospice services for many patients and their families. Patients should be referred to hospice when they become eligible in order to receive the full potential benefit from hospice in terms of quality of life and symptom management. More than this, for our hospice over 60% of our patients are currently served for 30 days or less. If the PAC provider waits for the 30 days to refer the patient, it would be logical that over 60% of the patients would die prior to hospice service. Over 35% of our patients are served for 7 days or less currently. Our median length of service is 15 days. Clearly, delays in referrals during the 30 day period would be devastating to both hospice and needy patients and families. This concern is further validated by the current skilled nursing facility situation for terminally ill Medicare patients discharged from the hospital. Skilled nursing facilities are able to qualify patients for Medicare reimbursement if the patient has had a hospital stay of three days or more, technically for rehabilitation although it is difficult to understand what rehabilitation is appropriate or reasonable for a patient with a life expectancy of less than six months, or 15 or 20 days. We find the skilled nursing facilities holding the referrals until the Medicare reimbursement has lapsed, the patient dies before the Medicare reimbursement lapses, and the patient and family are robbed of hospice services.

o Hospices provide bereavement services for families for thirteen months following the death of the patient. Bereavement support has been proven to improve the functioning and health of bereaved family members. Given that over 75% of hospice patients are over 75 years of age and almost 50% are over 85 years of age, many of these family members are also Medicare patients. Failing to provide bereavement services for these families will ultimately increase Medicare expenditures for these patients as well.

It is essential and appropriate that hospice be added to the list of service exclusions for each PAC provider. Making hospice a service exclusion for the MSPB-PAC measure will ensure that patients and their families will receive the right care at the right time at end-of-life. It will save Medicare dollars and provide for better health for the families after the death as well.

Please feel free to contact me if you have any questions. I look forward to the final MSPB – PAC Resource Use Measures.


COMMENT 39 OF 45

Date: 2/5/2016



Name, Credentials, Organization, and Email of Commenter: J. Donald Schumacher, PsyD,

President and CEO, National Hospice and Palliative Care Organization [email protected]


TEXT OF COMMENT: The National Hospice and Palliative Care Organization (NHPCO) appreciates the opportunity to submit comments on the draft specifications for the Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) Resource Use Measures, posted to the CMS website in January 2016. NHPCO is the largest membership organization representing the entire spectrum of not for profit and for profit hospice and palliative care programs and professionals in the United States. We represent over 4,000 hospice locations and more than 61,000 hospice and palliative care professionals in the United States, caring for the vast majority of the nation’s hospice patients. The organization is committed to improving end-of-life care and expanding access to hospice care with the goal of creating an environment in which individuals and families facing serious illness, death, and grief will experience the best that humankind can offer.

Hospices frequently care for patients after discharge from other post-acute care (PAC) providers and consequently have developed ongoing constructive relationships with many PAC providers to ensure continuity of care and optimize the transition to hospice care. Through our work in supporting hospices in developing and fostering positive relationships with PAC providers, NHPCO has gained an understanding of the PAC environment and the issues inherent in providing coordinated and efficient patient care in other post-acute settings. We also appreciate the importance of resource utilization measures for PAC providers – and believe that it is equally important that those measures accurately reflect the relationship between hospice and these providers.

Episode of Care and Associated Services

The proposed MSPB-PAC measure specifications describe an episode of care to include “the period a patient is directly under a PAC provider’s care, as well as a defined period after the end of that PAC provider’s treatment which may be reflective of and influenced by the services rendered by the PAC facility.” Associated services are Medicare Part A or Part B services, other than those in the treatment period, and are counted towards the episode cost in calculating the MSPB-PAC measure. The associated services period is 30 days of care after the discharge from a PAC provider.

Hospice is included under associated services in the proposed MSPB-PAC measure specifications and, therefore, costs for hospice care for the first 30 days after discharge from a PAC provider would be included in the calculation for resource utilization for the PAC provider.

NHPCO has reviewed the specifications for the proposed PAC resource utilization measures, including the methodology for construction of episodes of care and the designation for hospice utilization as a part of the episode. NHPCO has serious concerns that the inclusion of hospice as an associated services provider may result in patients who are eligible for hospice and could benefit from hospice care being denied timely and appropriate access to hospice services.



Disincentives for Hospice Referral

The MSPB-PAC measure creates a disincentive for referral to hospice by PAC providers because these providers are likely to be concerned that the cost of hospice will have a negative impact on their performance on the measure. This disincentive will result in denial of access to hospice services through a variety of pathways:

• PAC providers may not refer patients who are terminally ill to hospice and will either continue to care for them or simply discharge them. Hospice provides specialized, quality, compassionate care for people facing a life-limiting illness. The hospice interdisciplinary team is uniquely equipped to deliver expert medical care, pain management, and emotional and spiritual support expressly tailored to the patient's needs and wishes. Other PAC settings are not able to offer the same level of expertise and holistic comprehensive end-of-life care that hospice provides. PAC provider failure to refer patients to hospice because of concerns related to the MSPB-PAC measure will result denial of access to quality end-of-life care for many patients.

• PAC providers may wait to refer patients to hospice until death is imminent. While hospice providers are able to provide quality care for patients admitted to hospice services close to death, a very short length of service often means than patients and their families are not able to receive all of the available benefits of hospice care. The Medicare hospice benefit was designed to provide comprehensive, interdisciplinary care for beneficiaries in the last six months of life and their families. When hospice care is initiated close to death, patients and families are often in crisis. And while these patients and families still greatly benefit from hospice care, they could benefit even more from a longer length of service – particularly from psychosocial and spiritual care which utilizes therapeutic relationships and volunteer services that develop over time.

• PAC providers may also delay referral to hospice until after the 30 day associated services period has elapsed. A 30 day delay would deny timely access to hospice services for Medicare beneficiaries and their families. Patients should be referred to hospice when they become eligible in order to receive the full potential benefit from hospice in terms of quality of life and symptom management.

Service Exclusions

The specifications for the MSPB-PAC measures include a category of services that allow for some services to be excluded.

“Certain services are excluded from the treatment and associated services periods because they are clinically unrelated to PAC care and/or because PAC providers may have limited influence over certain Medicare services delivered by other providers during the episode window.”

“Inclusion of services that cannot be reasonably managed by the PAC provider could create incentives for providers to avoid treating patients with certain conditions or complex care needs … that cannot be fully accounted for in risk adjustment models.”

“Further, including such services in the MSPB-PAC measures would limit the extent to which


MSPB-PAC measures reflect meaningful and actionable variation in PAC providers’ efficiency.”

Section 3.1.5, pp 15-16

NHPCO believes that hospice services meet the above criteria for exclusion from the associated services period. And, furthermore, that inclusion of hospice as an associated service does not support the goal for MSPB-PAC measure to encourage higher quality PAC treatment in order to lower post-treatment costs.

The election of hospice care signifies shift in priorities and goals of care that is very different from care from other post-acute providers. Unlike other medical care, the focus of hospice care isn't to cure the underlying disease, meet functional status goals, or extend life. The goal of hospice care is to support the highest quality of life possible based on individual needs, that honors patients’ personal choices regarding care and addresses the issues most important to the patient and their family. The distinctive shift in focus and unique nature of hospice care makes hospice services clinically unrelated to PAC care.

In addition, patients at the end-of-life and their families have unique physical, emotional and spiritual needs that occur in the final phases of a life-limiting illness. Hospice is designed to meet those needs through specialized care that focuses on comfort measures; pain and symptom management, emotional support, spiritual support and quality of life. Each patient’s care is delivered through a team approach by an interdisciplinary team that includes nurses, home health aides, social workers, therapists, bereavement counselors, dieticians, volunteers, hospice chaplains, attending physician and hospice medical director utilizing an individualized plan of care. Hospice care providers offer specialized knowledge and support at the end-of-life that comprises a set of services that cannot be provided by the PAC provider.

Lastly, the distinctive focus of hospice care on comfort and maximization of quality of life together with the highly specialized nature of hospice care represent a significant difference from services provided by other PAC providers and hospice. That difference in the care provided by PAC providers and hospice providers limits the ability of the MSPB-PAC measures to reflect meaningful and actionable variation in PAC providers’ efficiency.

NHPCO strongly recommends that hospice be added to the list of service exclusions for each PAC provider. Designating hospice as a service exclusion for the MSPB-PAC measure will ensure that Medicare beneficiaries and their families will receive the right care at the right time at end-of-life.


COMMENT 40 OF 45

Date: 2/5/2016



Name, Credentials, Organization, and Email of Commenter: Troy Hillman, Director, PAC

Strategy and Analysis, Uniform Data System for Medical Rehabilitation [email protected]

Type of Organization: PAC data provider

TEXT OF COMMENT:

UDSMR would like to thank CMS and Acumen, LLC for this opportunity to comment on the “Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) Resource Use Measures”. As a provider of post-acute care assessment data collection, reporting and consultation services to over 900 post-acute providers, we appreciate the time and effort being put into the development of IMPACT Act measures, and look forward to future opportunities to provide feedback into this process. The current comment period specifically asks for comments related to the "overall episode construction methodology, exclusions, and the risk adjustment approach". UDSMR would like to offer the following comments and potential recommendations for your consideration: Overall Episode Construction Methodology: UDSMR is concerned that the current episode construction methodology may limit access to potentially necessary services and will inappropriately benefit "downstream" PAC providers who care for patients as the second or third PAC provider in a continuum of care. In these circumstances, the MSPB value for the initial or prior PAC provider will include spending amounts for associated services provided by the next PAC provider, while the next PAC provider will only have to account for their spending and any associated costs following their services. Will a potentially increased spending amount for the initial PAC provider cause the patient to be discharged home without receiving continued services provided in the existing continuum of care? Will medical or functional benefits made by the initial PAC provider allow the secondary provider to bear less risk of readmission or less potential for additional services post-discharge due to improved patient condition, and therefore provide the secondary provider with an increased opportunity to reduce their episode spending for that patient. To provide an example of this issue, our IRF database suggests that for the last 12 months roughly 15% of all IRF Medicare Fee for Service discharges are to a SNF level of care, while another 50% of all IRF Medicare Fee for Service discharges are Home with Home Health services. Roughly 65% of all IRF Medicare Fee for Service discharges will include in their IRF MSPB value the spending attributable not just to the IRF services, but also those services provided at the SNF or Home Health level for the 30 days after the IRF treatment period ends. The SNF or Home Health provider MSPB value will then include the spending attributable to their services and any associated services for the 30 days after their treatment period ends, but the SNF or Home Health provider does not include the IRF or other PAC spending that occurred prior to their treatment period, even though the prior PAC provider may have produced benefits to the patient that provide the SNF or Home Health provider with less risk of additional spending. Using data for 2012 discharges from the March 2014 MedPAC Medicare Payment Report to Congress, IRF average spending per patient is $17,995, SNF average spending per patient is $11,958 (based upon $28.7 billion spent on 2.4 million stays), and Home Health average spending per patient is $2,687. In an example where an IRF patient is discharged to a



SNF for additional services, the IRF MSPB value would be calculated based upon the IRF and SNF spending, which from the MedPAC data provided could be a total of $29,953. Meanwhile, the SNF MSPB value would be calculated on the SNF spending and any additional services for the 30 days post-discharge from the SNF, which from the MedPAC data could be $11,958 plus any additional service spending. Or for a SNF patient discharged to Home Health, the SNF MSPB value would include SNF and Home Health spending and could potentially be a total $14,645, while the Home Health MSPB value could utilize only $2,687 plus any additional service spending. If the measure truly is trying to capture the Medicare Spending per beneficiary in Post-Acute Care, shouldn't the secondary provider in these situations also have to report a value based upon the prior PAC setting spending, since post acute services were provided to this patient that resulted in their ability to limit or reduce the risk of their own spending on this patient? Since the initial PAC provider will essentially bear the burden of the secondary PAC spending, could this cause the initial PAC providers to consider limiting their referrals although these additional services may provide additional benefits to the patient as part of the continuum of care? To account for this issue, we would recommend or strongly urge that the measure developers consider including not only the spending for the 30 days after the end of the treatment period, but also any prior PAC spending for this patient that occurred prior to admission to the PAC provider being measures. This will then level the playing field for each type of provider who is caring for a patient within the continuum of care who may require medically necessary services from multiple PAC providers, and create a more robust measure of PAC spending per beneficiary that can be comparable between PAC venues. Exclusions: UDSMR appreciates the time and consideration the measure developers have placed into defining a category for services that may be excluded from the measure calculations. However, not enough detail is provided as to how providers may be able to determine what actual services may or may not be excluded. For example, will the spending for a potential acute care admission for influenza be excluded if a patient refuses the influenza vaccination in the prior PAC stay? Additionally, the documentation provided suggests that the lists of clinically assessed service exclusions are still being refined. Until such a time as these lists are finalized, how can any PAC provider be confident that their MSPB values will be measured in a standardized manner that will not vary over time due to the addition or subtraction of excluded services over time? We strongly recommend that the measure developers provide the lists of clinically assessed service exclusions so that PAC providers may be able to identify and potentially treat for those circumstances which may impact their resulting spending values. Risk Adjustment Approach: UDSMR recognizes that the measure developers have included a category or grouping for Prior PAC in the risk adjustment approach, however the approach does not define how this category may risk adjust or factor in the scenarios described above in our comments related to the Episode Construction Methodology. The approach indicates that the groupings will allow "for a more


accurate estimation of risk-adjusted spending for these sub-populations", but it does not indicate that it will actually factor in the spending from the prior PAC provider(s). Additionally, while the inclusion of HCC indicators may allow for consideration of medical factors, the measure developers fail to state when the 90-day look-back period begins, or whether diagnoses noted during the PAC treatment episode or within any of the associated services in the 30 days following the treatment episode will be eligible for HCC consideration. We strongly recommend that all clinically relevant diagnoses that are identified in the 90-day look back period, during the PAC treatment episode, or within any of the associated services in the 30 days following the treatment episode be considered for HCC risk adjustment. We also note the absence of any functional assessment data in the risk adjustment approach. While we recognize that a standardized functional data set has not yet been implemented for all PAC venues, we note that the various PAC payment systems utilize function as a basis for the payment amounts that will contribute to the MSPB values. We recommend that the measure developers consider inclusion of functional status in their risk adjustment approach where it is applicable utilizing existing assessment tools, with consideration given for modifying the risk-adjustment approach should standardized functional assessment data provide further risk-adjustment opportunities once fully implemented. Other Comments: UDSMR would appreciate clarification related to the use of average values in the numerator of this calculation, while the denominator utilizes a median value. The utilization of a median value would suggest that the underlying data may be skewed or have outlier values that may influence an average value. And since the median is used in the denominator for the MSPB-PAC Amounts for all providers in the same setting, could one question whether the underlying data utilized in calculating the MSPB-PAC Amounts may also be skewed or prone to outliers, such as the component of average standardized episode spending across all PAC providers of the same type? We question why the approach doesn't utilize mean or median values throughout the calculation. As previously stated, on behalf of UDSMR and the over 900 PAC subscribers we support we appreciate this opportunity to provide feedback on the MSPB measure and look forward to additional opportunities to contribute to the quality measures being developed for the IMPACT Act. Please let us know if we may be able to supply any additional information related to our comments or recommendations. Thank you for your time and consideration.


Regarding your comment on overall episode construction methodology, the measures include Medicare spending for services in the post-treatment period. In this way, providers are incentivized to deliver care to minimize the likelihood of the beneficiary requiring post-treatment services. This includes making the necessary referrals following the end of the treatment period,


without which the patient may relapse and incur higher cost services such as rehospitalization or emergency room usage.

Regarding your comment on clinical case mix categories, they do not include the spending from a prior PAC claim; rather they indicate the beneficiary's most recent institutional claim to improve the predictive power of the risk adjustment model. Please see Section 2.3.2 above generally for a discussion of risk adjustment issues.



Name, Credentials, Organization, and Email of Commenter: Donald L. Pendley, MA, CAE,

CFRE, APR, President, New Jersey Hospice and Palliative Care Organization

[email protected]


TEXT OF COMMENT:

Established in 1979, New Jersey Hospice and Palliative Care Organization (NJHPCO) represents 35 hospice providers in New Jersey and the 40,000-plus terminally-ill patients (and their families) they serve each year.

We read with interest the draft MSPB-PAC draft specifications and have concluded that including hospice as an associated services provider may result in patients being denied their federally-guaranteed right to the Medicare Hospice Benefit and the high-quality end-of-life services it provides.

Including hospice as an associated services provider would further delay referrals to hospice, already occurring at a startlingly late point. Over one-fourth of New Jersey’s hospice patients are on service for less than five days. Almost half are on service for less than two weeks.

The MSPB-PAC measure creates a disincentive for referral to hospice by PAC providers, who are likely to be concerned that the cost of hospice will have a negative impact on their performance in the measure. This disincentive will result in denial of access to hospice services for many reasons:

1. PAC providers might not refer patients who are terminally ill to hospice and will either continue to care for terminally-ill patients themselves or simply discharge them. This denies American citizens their federally-guaranteed right to high-quality end-of-life care. Dying is not merely a medical issue – it has significant social, emotional, psychological and spiritual impact, none of which are addressed by PAC providers as well as they are by interdisciplinary hospice teams. To get the fullest benefit from hospice’s psychosocial, spiritual and family-support services – and to reduce the rate of rehospitalization of terminal patients – hospice patients should be referred much earlier than many are now.



2. PAC providers may also delay referral to hospice until after the 30 day associated services period has elapsed. Such a delay would deny timely access to hospice services and patients and their families. Patients should be referred to hospice when they first become eligible in order to receive the full potential benefit from hospice in terms of quality of life and symptom management.

The specifications for the MSPB-PAC measure include a category of services that allow for some services to be excluded. “Certain services are excluded from the treatment and associated services periods because … PAC providers may have limited influence over certain Medicare services delivered by other providers during the episode window.” “Inclusion of services that cannot be reasonably managed by the PAC provider could create incentives for providers to avoid treating patients with certain conditions or complex care needs … that cannot be fully accounted for in risk adjustment models.” [Section 3.1.5, pp 15-16]

NJHPCO believes strongly that hospice services qualify for exclusion from the associated services period. Hospice patients have complex care needs. NJHPCO is concerned that the MSPB-PAC measure creates a disincentive for PAC providers to refer to hospice if hospice is categorized as an associated service and therefore part of the episode of care window.

We appreciate your attention to our comments and ask that the draft measures be modified to exclude hospice as an associated service, for the benefit of the millions of Americans who seek to be with their families in their dying time, and to be spared the burden of pain and debilitating symptoms.




Name, Credentials, Organization, and Email of Commenter: Frank Grosso, RPh, Executive

Director & CEO, American Society of Consultant Pharmacists [email protected]


TEXT OF COMMENT:

The American Society of Consultant Pharmacists (ASCP) is the only international professional society devoted to optimal medication management and improved health outcomes for all older persons. ASCP’s senior-care consultant pharmacist members manage and improve quality of life of geriatric patients and other individuals residing in a variety of environments, including nursing facilities, sub-acute care and assisted living facilities, psychiatric hospitals, hospice programs, and home and community-based care.



ASCP is pleased to offer comments on the Draft Specifications for the Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) Resource Use Measures.

Consultant pharmacists are licensed healthcare professionals that coordinate pharmacy services to optimize medication management for residents with facility staff, prescribers, and family/caregivers. Facilities contract with a consultant pharmacist to provide services that address the needs of each resident by conducting a Medication Regiment Review (MRR) at least monthly (State Operations Manuals F428), as well as assisting in facility-level issues such as medication errors, medication storage, education of staff, and other quality initiatives to improve medication use and safety. In addition to the requirement for monthly MRR (F428), the Medicare Part D final rule, issued in 2005 requires Part D plans to offer Medication Therapy Management (MTM) services. While similar, and having similar components, these services are different and both contribute to cost savings. These comments will outline MTM and MRR services, Part D population makeup, and how these services, broadly, lower the per beneficiary cost in the post-acute care setting.

MRR and MTM Services in CMS Regulation:

The current definition of Medication Regiment Review (MRR) in the State Operations Manual (SOM clearly acknowledges the collaborative work of the team, which is important in all care settings. The SOM definition states, “MRR is a thorough evaluation of the medication regimen of a resident, with the goal of promoting positive outcomes and minimizing adverse consequences associated with medication,” and the definition of MRR “…is a thorough evaluation of the medication regimen of a resident, with the goal of promoting positive outcomes and minimizing adverse consequences associated with medication. The review includes preventing, identifying, reporting, and resolving medication-related problems, medication errors, or other irregularities, and collaborating with other members of the interdisciplinary team.” (Adapted from the American Society of Consultant Pharmacists (ASCP) Guidelines for Assessing the Quality of Drug Regimen Review in Long--‐‑Term Care Facilities. Centers for Medicare and Medicaid Services, Electronic Health Record Incentive Program, Definition of Terms ‑ Medication Reconciliation: Updated May 2014. Retrieved September, 282015: https://www.cms.gov/Regulations--‐‑and--‐‑Guidance/Legislation/EHRIncentivePrograms/downloads/7_Medication_Reconciliation.pdf. “ Definition of Terms – Medication Reconciliation” – The process of identifying the most accurate list of all medications that the patient is taking, including name, dosage, frequency, and route, by comparing the medical record to an external list of medications obtained from a patient, hospital, or other provider.)

Additionally, CMS included MTM programs in the Medicare Part D final rule issued in 2005 (42 CFR 423.153 Drug Utilization Management, Quality Assurance, and Medication Therapy Management Programs). In this rule, CMS requires Part D Plans to include MTM programs that would manage the number, quality, and clinical necessity of beneficiaries’ medications. MTM services are designed to be a comprehensive education program for the beneficiary based on t heir condition(s), medications, and needed support services. MTM programs save money via patient-focused clinical education and streamlining medications for duplication, effectiveness, and potential adverse interactions and outcomes.

https://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/downloads/7_Medication_Reconciliation.pdf


Changing Models in Post-Acute and Long-term Care

Current Medicare incentives to improve patient satisfaction, reduce cost, and improve care (the “Triple Aim”), has led to a decline in nursing facility census levels which are at the lowest point in 10 years (http://www.ahcancal.org/research_data/trends_statistics/Documents/Trend_PVNF_FINALRPT_June2014.pdf)

The Centers for Medicare & Medicaid Services reported in July 2015 that the number of Medicare patients who receive some skilled nursing care annually increased from 636,000 (19 per 1,000 enrollees) in 1989 to 1,839,000 (52 per 1,000 enrollees) in 2010. This data indicates that the old model of long-stay nursing homes is shifting. More patients are using skilled nursing facilities for short-stay and post-acute care rehabilitation, and returning home.

Regardless of the length of stay, on average, these patients are 65 years of age or older and are moving through long-term facility-based care settings. They tend to have many severe chronic and co-morbid conditions and for which they are prescribed multiple medications. Seniors represent just over 13% of the population, but consume 40% of prescription drugs and 35% of all over the counter drugs. According to a 2009 Alliance for Aging Research report, on average, individuals 65 to 69 old take nearly 14 prescriptions per year, individuals aged 80 to 84 take an average of 18 prescriptions per year. (https://www.ascp.com/sites/default/files/file_Task_Force_2009_FINAL-3.pdf ) Adverse drug reactions are responsible for 32,000 hip fractures and 28 percent of re-hospitalizations; patients 65 and older are two-and-a-half times more likely to visit the emergency room because of an adverse drug reactions [sic] than patients in the under-65 population. (ibid)

Adverse outcomes from medication mismanagement represents one of the leading causes of re-hospitalization after discharge. By consulting with a patient before leaving the hospital, senior care pharmacists can dramatically reduce post-acute care costs. According to the Henry J. Kaiser

http://www.ahcancal.org/research_data/trends_statistics/Documents/Trend_PVNF_FINALRPT_June2014.pdf

https://www.ascp.com/sites/default/files/file_Task_Force_2009_FINAL-3.pdf


Family Foundation, in 2014, the average cost of a one-day hospital stay was $2,212 (ranging from a low of $1,331 in Wyoming to $3,344 in Oregon).

Identifying Chronically Ill Populations and Improving Quality of Care

The prevalence of multiple chronic conditions and functional impairment within the aging population is increasing. Today, the 15 percent of Medicare enrollees with both chronic conditions and functional limitations who need long-term services and supports account for one-third of Medicare spending. Care coordination can help to improve clinical outcomes for this population.

The increasing prevalence of older adults with functional status deficiencies requires trained clinicians, like pharmacists to take steps to determine the most efficient and most clinically appropriate treatment plans. ASCP encourages the report to reflect importance of functional limitations in Part D when evaluating per-beneficiary spending.

Savings With MRR and MTM:

Although Medication Regimen Review and Medication Therapy Management have differing clinical focus and different audiences (MRR recommendations directed to clinical staff, MTM directed to the patient), both are potent tools in improving clinical outcomes and reducing health system expenditures. In both MTM and MRR, the pharmacist closely examines all medications the patient is using, including prescribed drugs and OTC/dietary supplements. Lab work, care plans, and clinician documentation are scrutinized in the MRR process; in MTM, the patient or patient caregiver is interviewed directly with respect to medication usage. Often, inconsistencies, including medication redundancies (polypharmacy), potential drug interactions, inappropriate medications, and poor adherence due to medication intolerance, are discovered. Ensuing recommendations generally result in an improved, patient-specific medication regimen that often requires fewer agents, and is more tolerable to the patient. In the MTM process, the pharmacist designs a regimen that factors in medication-specific protocols, such as whether a drug should be ingested with food or on an empty stomach, the most appropriate time of day to take drug [sic], and possible adverse events that should be reported back to the prescriber. This scrutiny of medication regimens, whether in a long term care facility during the MRR process, or directly with a patient during a MTM encounter, is particularly critical with seniors, who often see many different clinical specialists, resulting in high potential for unnecessary expenditures and medication misadventures.

MTM and MRR, among other services offered by consultant pharmacists in the post-acute care settings go a long way to keep the Medicare per-beneficiary costs low by preventing costly re-hospitalizations.

RESPONSE: Thank you for your comment. We appreciate your interest and concern in the role that medication management programs can play in limiting Medicare spending. The MSPB-PAC measures however do not include Part D drugs. Please see Summary Comment ID-23 for a discussion of Part D.




Name, Credentials, Organization, and Email of Commenter: Dianne Hansen, MT, MHA,

BCHH-C, COS-C, Director of Clinical Operations, Partners In Home Care

[email protected]

Type of Organization: Hospice

TEXT OF COMMENT:

I appreciate the opportunity to comment on the proposal published by the CMS contractor, Acumen, related to Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) Resource Use Measures. CMS has asked specifically for comment from hospice stakeholders on these measures.

The MSPB-PAC measures are designed to provide data for the purpose of improved efficiency and coordination of care in the post acute care (PAC) setting. The ultimate effect will be that PAC providers will be accountable for resource utilization during the episode of care, including services for 30 days after discharge from a PAC provider. In the proposal set out by Acumen, hospice is considered an “associated service provider,” and as such hospice services would be considered part of the resource utilization for the PAC provider.

As a hospice provider, our agency has serious concerns about including hospice as an associated service provider. Such structure gives rise to concerns that PAC providers will not refer to hospice, or they will refer patients late in their disease. It has been our experience that patient and families are most satisfied with hospice care when patients are referred earlier in their disease decline and have opportunity to take full advantage of all the services that the hospice benefit was designed to offer, including nursing, social work, spiritual counseling, expert pain and symptom control, volunteers, and bereavement services. Families whose loved one was admitted close to death with a very short length of stay often feel that they are less prepared for what to expect from the death process, giving rise to complicated grief issues after the patient passes.

Classifying hospice as an associated service provider would increase the risk that PAC providers may (1) wait to refer to hospice until death is imminent, (2) delay referral until after the 30-day associated service period has passed, or (3) not refer patients to hospice.

The specifications for the MSPB-PAC measures allow for some services to be excluded (Section 3.1.5, pp 15-16). Our agency believes that hospices services meet this exclusion criteria due to the fact that PAC providers “may have limited influence over certain Medicare services” and “inclusion of services that cannot be reasonably managed by the PAC provider could create incentive for providers to avoid treating patients with certain conditions or complex care needs . . . ” Hospice patients do have very complex needs, and the acuity of hospice patients has risen significantly over past years. Hospice providers are best equipped to deliver the expert medical care, pain management, and emotional and spiritual support to meet a hospice patient’s goals.



We strongly recommend that hospice services be added to the list of service exclusions for PSPB-PAC measures.[sic]




Name, Credentials, Organization, and Email of Commenter: Annette Kiser, Director of

Quality & Compliance, The Carolinas Center [email protected]


TEXT OF COMMENT:

The Carolinas Center appreciates the opportunity to submit comments on the draft document, “Medicare Spending Per Beneficiary – Post-Acute Care (MSPB-PAC) Resource Use Measures,” published by CMS contractor, Acumen. The Carolinas Center is the only state association exclusively focused on serving hospice and palliative care providers in North Carolina and South Carolina. We represent an extensive network of Medicare-certified hospice programs, many with multiple locations, which provide care to thousands of patients every day. Our goal is to assist providers with fulfilling their missions of delivering exceptional care to patients at the end of life as well as to the patients’ caregivers.

The proposed specifications address measures for reporting of data in post-acute care (PAC) settings: skilled nursing facilities (SNFs), home health agencies (HHAs), long-term care hospitals (LTCHs), and inpatient rehabilitation facilities (IRFs). CMS has requested comments from stakeholders on the treatment of episodes with hospice and that is the focus of our comments.

The Carolinas Center (TCC) understands the need to encourage efficiency and coordination in PAC settings, but have concerns with including hospice services in the list of “associated services” that will impact the post-treatment period measure calculation. The median length of service for hospice patients in 2014 was less than 18 days. Any action that can further shorten this length of service will be detrimental to patients and families at their most vulnerable time – the end of life.

The Carolinas Center offers the following comments on the quoted language from the draft document:

• “Given this design of the MSPB-PAC measures, post-treatment costs may serve as an indicator of the quality of care provided during PAC care, in that higher quality PAC treatment may yield lower post-treatment costs.”



o Since hospice care is provided at a fixed per diem, the PAC provider would not impact hospice spending after the PAC episode ends and this could cause concern for the PAC provider. We offer that referring patients to hospice earlier in the PAC phase can lead to better patient outcomes and overall reduced costs for the Medicare system because the focus would shift to palliative care to manage pain and symptoms as opposed to aggressive treatment of medical conditions.

• “As a beneficiary moves from one provider to the next in his/her care trajectory, every PAC and hospital provider that the beneficiary encounters will have incentives to deliver cost efficient care.”

o This statement implies that it is in the best interest of the PAC provider to reduce the utilization of other services. The hospice per diem would add approximately $4500 to the 30-day period for associated services. PAC providers would want to avoid this cost and may delay referrals to hospice until the associated services period ends.

• Treatment services are defined in part as services that are “necessary to maintain a patient’s health status or assist in recovery (e.g., evaluation and management, screening).”

o SNF, HHA, LTCH, and IRF services are not Medicare-reimbursed services that are used by the majority of hospice patients; so, there is little relation to the PAC treatment episode.

o It is more likely that any utilization of the PAC services would be directed to rehabilitation for the patient, or to maintain the health status. When a patient’s condition declines and they are no longer able to continue the rehabilitative or maintenance focus, hospice admission would be indicated. There should be no disincentives, real or perceived, for the PAC provider to refer a patient to hospice.

• The draft specifications state “Certain services are excluded from the treatment and associated services periods because they are clinically unrelated to PAC care and/or because PAC providers may have limited influence over certain Medicare services delivered by other providers during the episode window. Inclusion of services that cannot be reasonably managed by the PAC provider could create incentives for providers to avoid treating patients with certain conditions or complex care needs (e.g., patients requiring chemotherapy or dialysis) that cannot be fully accounted for in risk adjustment models.”

o Hospice services would meet the exclusion criteria in that the PAC provider cannot manage the hospice services because hospice services are prescribed by statute and Medicare policy and must be delivered as a total benefit paid on a per diem basis.

With the short length of service for hospice patients, any further delays in admission to hospice will serve only to perpetuate the problem of patients and families receiving care at the brink of death. To maximize the benefit of hospice services, the interdisciplinary team needs time to develop a rapport with the patient and family, to provide education and to coordinate all necessary resources to meet the needs of the patient and those involved in their care.

The Carolinas Center requests that hospice be excluded from the associated services category for the measures for all post-acute care providers.


We appreciate this opportunity and are available to provide additional information as CMS moves forward with the measures specification process.




Name, Credentials, Organization, and Email of Commenter: Kathryn Brod, President/CEO,

LeadingAge Ohio [email protected]


TEXT OF COMMENT:

LeadingAge Ohio is grateful for the opportunity to provide comment on the development of measures that will enable comparisons between post-acute care providers, as required in the Improving Medicare Post-Acute Care Transformation (IMPACT) Act of 2014. LeadingAge Ohio is a state affiliate of LeadingAge, representing mission-driven providers of pre- and post-acute care, services and supports. Our membership consists of continuing care retirement facilities, skilled nursing facilities, assisted living centers, adult day providers, housing organizations, and home health and hospice agencies in the state of Ohio.

LeadingAge Ohio is supportive of the motivation behind the IMPACT Act: to standardize information capture and reporting to allow for mutual comparability of Medicare post-acute care providers. We are concerned, however, about the current proposal as well as the very meager amount of time that has been allowed for the provider industry to fully digest and react to the proposals at hand. We are pleased that the deadline for comments has been extended twice, stretching the comment period from just two weeks to four, but this is still a very, very short timeframe for providers and their representative associations to fully consider the possible scenarios and fallout from the proposals at hand.

Specifically related to the MSPB-PAC, there are six points that we raise for your consideration.

• LeadingAge Ohio takes exception to these measures being described as “efficiency” measures. Efficiency is typically conceptualized as dependent on two contributing factors: value and cost. The measures as proposed only capture cost, and do not incorporate any measurement of the outcomes gained for that cost. As such, they should be referred to as measures of cost, expense, or resource use.

• LeadingAge Ohio encourages Acumen to consider excluding hospice care from the associated services which are incorporated into calculating these measures. Our specific concern is that including hospice as an associated service would create disincentives for timely hospice referral. Including hospice in the “associated services” category would mean that any PAC provider monitoring its performance on this measure would be



concerned with resource use in the thirty days following discharge. This means that there would only be an incentive for hospice referral for those patients who are so actively declining that they are likely to end up in a more intensive, more costly care setting 30 days following discharge. It would disincentivize referral for all those with uncertain prognosis, but which would clearly meet hospice eligibility requirements and be well served by hospice’s all-inclusive case management model. Beginning on January 1, hospices began to work under a new payment model which was designed to be more “incentive-neutral.” It would be imprudent to take a step backward by inappropriately designing these measures.

• LeadingAge Ohio believes that an accurate measure of Medicare spending should incorporate Medicare Parts A, B, and D expenditures, especially given the escalation of polypharmacy and how it leaves patients vulnerable to medical errors. Pharmaceuticals fall within the realm of influence for post-acute care providers, and prudent use of medications is a mark of a high-quality care provider. The omission of pharmaceutical expenditures should be reconsidered.

• LeadingAge Ohio believes the episode length used for standard home health and LUPA episodes should not be sixty days. For cases when the home health episode ends before day sixty, the treatment period should end on the day of discharge. Roughly nine percent of episodes and one percent of payments are from LUPAs, so while this is a small minority, it would be a more fair and meaningful way to capture these costs.

• Additionally, post-acute home health differs inherently from community-admitted home health care. For purposes of mutual comparability between post-acute providers, LeadingAge Ohio encourages Acumen to limit this calculation to that subset of home health care which is post-acute in nature. While this cuts the population captured significantly—roughly a third of home health episodes follow an acute stay—it is the only way for the data to be meaningful and meet the goals of the IMPACT act, which is to unify post-acute care across settings.

• LeadingAge Ohio supports risk-adjustment for individual characteristics, but believes that the proposal falls short by excluding social and economic factors, which are often related to resource use. Income level alone would provide significant information: individuals who are able to afford to purchase in-home support may be eligible for discharge to home sooner than an individual who does not have the means to pay. One would expect that patients with fewer means would have to rehabilitate to a higher functional level prior to being discharged home. Additionally, the Centers for Medicare and Medicaid Services are focusing on social determinants of health in racial and ethnic minority groups, and these, too, have a bearing on Medicare resource use in post-acute settings.

LeadingAge Ohio appreciates the opportunity to provide feedback on these important measures of post-acute care. Please do not hesitate to contact us if you have any questions.
